An adequate ADME/PK (absorption, distribution, metabolism, excretion, pharmacokinetics) profile is an essential quality for a drug. As part of the drug discovery process, leads are iteratively designed and optimized in order to simultaneously satisfy various properties such as appropriate ADME/PK levels and high biological activity for a target. The drug discovery process can be accelerated by improving the likelihood that a designed compound fulfils the necessary pharmacologic properties, and thus reducing the number of needed iterations. A promising technique is de novo drug design, where molecules are computationally generated based on a set of desired attributes. Our project aimed to benchmark the effectiveness of the ANDROMEDA ADME/PK conformal prediction models in guiding the generation of compounds toward an area of chemical space with good ADME/PK properties. For this, we used the REINVENT reinforcement learning framework built by the Molecular AI team at AstraZeneca. Here, we integrated 4 out the 14 available ANDROMEDA models (fabs , fdiss, CLint and Vss) as oracles in the scoring component of the generative model. Oral bioavailability (F) is a secondary parameter that was computed with the help of the aforementioned models and fu(unbound fraction in plasma), and serves as the fifth ADME/PK oracle in our analysis. We aimed to rediscover DRD2 bioactives with a good ADME/PK profile. Our results show that the ANDROMEDA models have a slight influence on the predicted ADME/PK properties of the generated compounds. The results do not show an increased likelihood of generating DRD2 ligands in the case of the primary ANDROMEDA models. However, when using the oral bioavailability oracle, the sampling likelihood increases for some of the approved DRD2 ligands. In conclusion, the oral bioavailability ANDROMEDA model can be a promising option for guiding the generation of novel compounds towards an area of chemical space with good ADME/PK properties.

There has been a desire to accurately interpret the inhaled pharmacokinetic (PK) profiles of drugs in humans to aid successful inhaled drug and product developments. However, challenges are layered, as 1) the drug dose delivered to the lung (DTL) from inhalers is a portion of the formulated dose but rarely determined; 2) lung delivery and regional deposition differ, depending on drug, formulation and inhaler; 3) drugs are not only absorbed from the lung but may also be from the gastrointestinal (GI) tract; and 4) in addition to absorption into the systemic circulation, multiple non-absorptive processes also eliminate drugs from the lung, such as mucociliary clearance, metabolism, phagocytosis and tissue binding. Hence, this thesis project aims to develop new lung disposition model-based analyses to derive the meaningful kinetic descriptors for lung disposition from inhaled PK profiles in humans. Two approaches, curve fitting- and moment-based approaches, were developed. Both approaches modeled the kinetics of lung disposition rate-controlled by absorption (ka) and non-absorptive loss (knal), assuming no contribution of GI absorption. An exhaustive literature review found necessary data sets for three drugs, tobramycin, calcitonin and ciprofloxacin. In the curve fitting-based approach, each inhaled PK profile was fitted to the lung disposition model, while the DTL was obtained from corresponding -scintigraphic lung deposition and the kinetic parameters of systemic disposition were fixed by separate intravenous PK profile model analysis. In the moment analysis-based approach, the mean lung residence times (MLRT) and the DTL-based bioavailability (FL) were estimated and used to determine the ka and knal values in the lung disposition model, given FL = MLRTka = ka/(ka+knal). The ka and knal values were successfully derived for all the three drugs delivered by dry powder inhalers (DPIs) and/or nebulizers (NEB) through both approaches. Their “goodness-of-fit” was reasonably satisfactory. The ka values appeared to be primarily described by partition-based diffusion affected by the three hydrophilic drug’s molecular weight. In contrast, the knal values differed, yet appeared to become plausible, with a notion of additional non-absorptive confoundedness due to lung tissue binding (tobramycin) and metabolism (calcitonin), in addition to mucociliary clearance. The ka and knal values derived by the two approaches were comparable in majority of the cases. The success of these PK modeling analyses enabled further attempts to identify most influential attributes by simulation. The systemic PK and lung exposure profiles were predicted by simulation upon ±20 % changes in each of the DTL, ka and knal values to examine changes in the systemic PK metrics (Cmax, AUC and Tmax) and local lung exposure metrics (AUClung and LRT0.5). For all three drugs, the Cmax and AUC changes were identical to changes in the DTL without changing the Tmax. In contrast, impacts of the ka and knal changes differed between drugs, depending on the relative contribution of the rate constant to their sum (ka+knal). It appeared that the major contributor of the sum (ka+knal) was that rate-controlling the kinetics of lung disposition. In conclusion, this thesis project has successfully proposed two new approaches of curve fitting and moment-based analysis by accurately deriving the kinetic descriptors of lung disposition (ka and knal) for three drugs from the inhaled PK profiles in humans. Their applications were extended to predict likely changes in the systemic PK and local lung exposure metrics by simulation. While attempts should continue with more drugs, these approaches are believed to be useful in identifying critical attributes to determine the lung disposition kinetics and thus predicting the lung kinetic behavior and systemic PK profiles of new drug entities in humans.

RNA nanotechnology is an emerging field that holds great promise for advancing drug delivery and materials science. Recently, RNA nanoparticles have seen increased use as an in vivo delivery system. RNA was once thought to have little potential for in vivo use due to biological and thermodynamic stability issues. However, these issues have been solved by: (1) Finding of a thermodynamically stable three-way junction (3WJ) motif; (2) Chemical modifications to RNA confer enzymatic stability in vivo; and (3) the finding that RNA nanoparticles exhibit low immunogenicity in vivo. In vivo biodistribution and pharmacokinetics are affected by the physicochemical properties, such as size, shape, stability, and surface chemistry/properties, of the nanoparticles being delivered. RNA has an inherent advantage for nanoparticle construction as each of these properties can be finely tuned. The focus of this study is as follows: (1) Construction of diverse size and shape RNA nanoparticles with tunable physicochemical properties; (2) Investigation of the effect that size, shape, and nanoparticle properties have on in vivo biodistribution; (3) Development of drug encapsulation and release mechanism utilizing RNA nanotechnology; and (4) Establishment of large-scale synthesis and purification methods of RNA nanoparticles. In (1), RNA triangle, square, and pentagon shaped nanoparticles were constructed using the phi29 pRNA-3WJ as a core motif. Square nanoparticles were constructed with sizes of 5, 10, and 20 nanometers. The RNA polygons were characterized by AFM to demonstrate formation of their predicted geometry per molecular models. Furthermore, the properties of RNA polygons were tuned both thermodynamically and chemically by substitution of nucleic acid type used during nanoparticle assembly. In (2), the biodistribution of RNA nanosquares of diverse sizes and RNA polygons of diverse shapes were investigated using tumor models in nude mice. It was found that increasing the size of the nanosquares led to prolonged circulation time in vivo and higher apparent accumulation in the tumor. However, it was observed that changing of shape had little effect on biodistribution. Furthermore, the effect of the hydrophobicity on RNA nanoparticles biodistribution was examined in mouse models. It was found that incorporation of hydrophobic ligands into RNA nanoparticles causes non-specific accumulation in healthy organs, while incorporation of hydrophilic ligands does not. Lower accumulation in vital organs of hydrophobic chemicals was observed after conjugation to RNA nanoparticles, suggesting RNA has the property to solubilize hydrophobic chemicals and reduce accumulation and toxicity in vital organs. In (3), a 3D RNA nanoprism was constructed to encapsulate a small molecule fluorophore acting as a model drug. The fluorophore was held inside the nanoprism by binding to an RNA aptamer. The ability of the stable frame of the nanoprism to protect the fragile aptamer inside was evidenced by a doubling of the fluorescent half-life in a degrading environment. In (4), a method for large-scale in vitro synthesis and purification of RNA nanoparticles was devised using rolling circle transcription (RCT). A novel method for preparing circular double stranded DNA was developed, overcoming current challenges in the RCT procedure. RCT produced more than 5 times more RNA nanoparticles than traditional run-off transcription, as monitored by gel electrophoresis and fluorescence monitoring. Finally, large-scale purification methods using rate-zonal and equilibrium density gradient ultracentrifugation, as well as gel electrophoresis column, were developed.

The rationalisation of drug potency using three-dimensional structures of protein-ligand complexes is a central paradigm in medicinal research. For over two decades, a major goal has been to find the rules that accurately relate the structure of any protein-ligand complex to its affinity. Addressing this problem is of great concern to the pharmaceutical industry, which uses virtual screens to computationally assay up to many millions of compounds against a protein target. A fast and trustworthy affinity estimator could potentially streamline the drug discovery process, reducing reliance on expensive wet lab experiments, speeding up the discovery of new hits and aiding lead optimization. Water plays a critical role in drug-protein interactions. To address the often ambiguous nature of water in binding sites, a water placement method was developed and found to be in good agreement with X-ray crystallography, neutron diffraction data and molecular dynamics simulations. The method is fast and has facilitated a large scale study of the statistics of water in ligand binding sites, as well as the creation of models pertaining to water binding free energies and displacement propensities, which are of particular interest to medicinal chemistry. Structure-based scoring functions employing the explicit water models were developed. Surprisingly, these attempts were no more accurate than the current state of the art, and the models suffered from the same inadequacies which have plagued all previous scoring functions. This suggests a unifying cause behind scoring function inaccuracy. Accordingly, mathematical analyses on the fundamental uncertainties in structure-based modelling were conducted. Using statistical learning theory and information theory, the existence of inherent errors in empirical scoring functions was proven. Among other results, it was found that even the very best generalised structure-based model is significantly limited in its accuracy, and protein-specific models are always likely to be better. The theoretical framework developed herein hints at modelling strategies that operate at the leading edge of achievable accuracy.

Bromodomains (BRDs) are protein modules which bind to acetylated lysines on histones and transcriptional regulating proteins. BRD-containing proteins are involved in a large variety of critical cellular processes and their misregulation, or mutation of the genes encoding for them, has been linked to pathogenesis in humans. The generation of chemical probes (potent, selective and cell permeable small molecules) in cellular experiments to investigate the biological role of the BRDs is thus desirable. A chemical probe for the CREB (cyclic-AMP response element binding protein) binding-protein (CBP) and E1A binding protein (p300) BRDs was developed, starting from a low molecular weight, weak and non-selective dimethylisoxazole benzimidazole compound. Parallel synthesis was used to optimise the initial hit into a weak, but selective CBP inhibitor. Further modification of the two N-1 and C-2 moieties of the benzimidazole scaffold, led to highly potent and selective CBP inhibitors. Structure-guided design was then applied to optimise the selectivity of the series for CBP over the first domain of bromodomain-containing protein 4 BRD4(1). A strategy to reduce the flexibility of the N-1 and C-2 ethylene linker groups through the incorporation of conformational constraints led to inhibitors with increased selectivity. The optimal compound was highly potent for the CBP and p300 BRDs (K_d 21 nM and 32 nM, respectively) and selective over BRD4(1) (40-fold and 27-fold, respectively). On-target cellular activity was observed in a fluorescence recovery after photobleaching (FRAP) assay (0.1 μM), a p53 reporter gene assay (IC₅₀ 1.5 μM) and a Förster resonance energy transfer (FRET) assay (5 μM). A weak indolizine bromodomain-containing protein 9 (BRD9) inhibitor was used as the starting point for the development of a BRD9/BRD7 chemical probe. Analogues were synthesised via [3+2] cycloadditions. An optimised compound was found to be highly potent (68 nM) and selective over BRD4(1) (34-fold). On-target cellular activity was observed in a FRAP assay (5 μM). Efforts were made to improve the cellular activity through the introduction of an ionisable centre to aid solubility. A selection of piperazine analogues were shown to be potent and selective, and these compounds warrant further investigation of their selectivity and cellular activity. Overall, the work has led to the first potent and selective inhibitors of the CBP/p300 and BRD9 BRDs. It also highlights the role of structural analysis in the development of inhibitors that modulate protein-protein interactions.

Lactate Dehydrogenase-A (LDH-A) is up-regulated in a broad array of cancers and is associated with poor prognosis. Involved in the hypoxic response, LDH-A is a HIF-1 target and is responsible for the enzymatic reduction of pyruvate to lactate. This is important for several reasons, chiefly (1) the regeneration of NAD+ which feeds back into earlier glycolytic stages and (2) the depletion of intracellular pyruvate concentrations. High intracellular pyruvate is known to inhibit HDACs and is associated with increased apoptosis. LDH-A is also known to be controlled by oncogenes such as c-Myc suggesting an oncogenic role. Studies have shown that the knock-out of LDH-A reduces proliferation and tumourgenicity, and stimulates the mitochondria. This thesis therefore had three aims: firstly, to validate LDH-A inhibition and elucidate its full nature in terms of the implications for tumour survival; secondly, to ascertain the role of LDH-B in order to determine whether selectivity towards LDH-A would be a necessary feature of any small molecule; lastly, to recapitulate siRNA mediated LDH-A inhibition with small molecule inhibitors that had the potential for clinical application. The thesis examined both clinical data and a broad panel of cultured cancer cell types in order to select appropriate model in which to validate siRNA mediated inhibition of LDH-A and LDH-B. After it was demonstrated that LDH-A inhibition reduced the growth of cultured cells, a range of techniques were used to quantify this reduced growth in terms of cell death and changes in metabolism. Further to this, literature studies had proposed a role for LDH-B in maintaining lactate fuelled tumour growth; however, this thesis shows that in the cell lines studied, lactate-fuelled tumour growth was an LDH-A dependent phenomenon. Finally, a high throughput assay system was designed and validated and a library of small molecules was selected, synthesized, and screened in order to identify selective inhibitors of LDH-A.

The purpose of the project was to investigate the impact of IV and PO routes difference for MDZ, a prototypical CYP3A substrate, and two CYP3A inhibitors (CYP3AI) -FLZ and ERY-, on the magnitude and time course of their inhibitory metabolic DDI. Individual semi-PBPK models for MDZ, FLZ and ERY were developed and validated separately, using pharmacokinetic (PK) parameters from clinical/in-vitro studies and published physiological parameters. Subsequently, DDI sub-models between MDZ and CYP3AIs incorporated non-competitive and mechanism-based inhibition (MBI) for FLZ and ERY, respectively, on hepatic and gut wall (GW) CYP3A metabolism of MDZ, using available in-vitro/in-vivo information. Model-simulated MDZ PK profiles were compared with observed data from available clinical PK and DDI studies, by visual predictive check and exposure metrics comparison. DDI magnitude and time course for CYP3AI (IV vs. PO) followed by MDZ (IV vs. PO) at various time points were predicted by the validated semi-PBPK-DDI models. Two hypothetical CYP3A substrates and four CYP3AI (derived from MDZ, FLZ and ERY, with GW metabolism removed, hepatic metabolism reduced, or oral bioavailability (Foral) and/or elimination half-life (t1/2) modified) were also simulated to generalize conclusions. The final semi-PBPK-DDI models predict well the PK profiles for IV/PO MDZ in absence/presence of IV/PO CYP3AI, with deviations between model-predicted and observed exposure metrics within 30%. Prospective simulations demonstrate that: 1) CYP3A substrates, e.g., MDZ, are consistently more sensitive to metabolic inhibition after PO than after IV administration, due to pre-systemic hepatic and/or GW metabolism. For substrates without GW metabolism and limited hepatic metabolism, only a marginal route difference for substrate administration is observed. 2) For high-Foral CYP3AIs, e.g., FLZ, no inhibitor IV-PO route DDI differences are expected, unless they are given simultaneously with PO MDZ. 3) For low-Foral CYP3AIs, e.g., ERY, greater inhibition is expected after IV than after PO administration for IV MDZ, but is difficult to predict for PO MDZ. 4) In addition to Foral and plasma t1/2 of CYP3AIs, the DDI onset, peak and duration are determined by their oral absorption rate and by the resulting hepatic and/or GW concentration profiles relative to Ki for noncompetitive CYP3AIs, but by CYP3A kinetics (synthesis, degradation rate) for MBI CYP3AIs.

Molecular modelling has become a well-established tool for studying biological mole- cules, moreover with the prospect of being useful for drug development. The thesis summarises research on the methodological advances in the treatment of molecular flexibility and intermolecular interactions. Altogether, seven original publications are accompanied by a text which aims to provide a general introduction to the topic as well as to emphasise some consequences of the computer-aided drug design. The molecular flexibility is tackled by a study of a drug-DNA interaction and also by an investigation of small drug molecules in the context of implicit solvent models. The approaches which neglect the conformational freedom are probed and compared with experiment in order to suggest later, how to cope with such a freedom if in- evitable. The noncovalent interactions involving halogen atoms and their importance for drug development are briefly introduced. Finally, a model for a faithful description of halogen bonds in the framework of molecular mechanics is developed and its per- formance and limits are tested by a comparison with benchmark ab initio calculations and experimental data. 1

A doença de Chagas, causada pelo Trypanosoma cruzi, é uma doença tropical que aflige milhões de pessoas, gerando consequências sócio-econômicas devastadoras. Ela tem sido considerada uma doença tropical super-negligenciada, já que os únicos fármacos disponíveis para o seu tratamento apresentam baixa eficácia e causam vários efeitos colaterais. Além disso, os mesmos foram introduzidos há mais de três décadas. Com esse cenário, é evidente a necessidade da descoberta, desenvolvimento e introdução de novos fármacos para o tratamento eficiente e seguro da doença de Chagas. A enzima gliceraldeído-3-fosfato desidrogenase (GAPDH) é um alvo biomacromolecular atraente para a descoberta de novos fármacos contra os tripanossomatídeos, em virtude das enzimas da via glicolítica exercerem um papel fundamental no fornecimento de energia para a sobrevivência do parasito. Essa enzima foi selecionada neste trabalho de tese para a realização de estudos em química medicinal com base em quiminformática com o objetivo de identificar potenciais inibidores enzimáticos e do T. cruzi. Na primeira etapa desta tese, o ensaio virtual baseado na estrutura do alvo (SBVS) foi usado na identificação e seleção dos compostos. Como resultado do planejamento in silico, vinte compostos foram selecionados e avaliados experimentalmente na segunda etapa do trabalho empregando a técnica de calorimetria de titulação isotérmica (ITC). Destes, onze inibiram a GAPDH de T. cruzi resultando numa elevada taxa de acerto (>= 20%). Os novos inibidores apresentam excelente eficiência do ligante (LE), bem como mostram ligeira seletividade pela enzima do parasito. O ensaio dos inibidores contra a forma tripomastigota do T. cruzi identificou dois compostos capazes de inibir essa forma infectiva e um deles também mostrou ser um potente inibidor da forma amastigota do parasito, além de apresentar baixa toxidez. As duas melhores classes de inibidores da GAPDH e do parasito foram selecionadas para o estabelecimento de relações quantitativas entre a estrutura química e a atividade biológica (QSAR). Estudos de QSAR 2D (HQSAR) forneceram modelos com elevada capacidade preditiva e proporcionaram a identificação de características estruturais importantes para a otimização dos ligantes a compostos-matrizes. / Chagas disease, caused by Trypanosoma cruzi, is a tropical disease, which afflicts millions of people, thus generating devastating socio-economic consequences. It has been pointed out that it is a super-neglected tropical disease, based on available drugs with low efficacy and that give rise to many side effects. In addition, these drugs were introduced three decades ago. With this scenario, it is clear the necessity of the discovery, development and introduction of new efficient drugs to treat Chagas disease. The enzyme glyceraldehyde-3-phosphate dehydrogenase is a promising target for the development of new drugs against trypanosomatides, since the enzymes of the glycolytic pathway display a fundamental role in the energy supply to parasite survival. In this thesis, this enzyme was selected for medicinal chemistry within the cheminformatics framework aiming at the identification of potential enzymatic and parasite inhibitors. In the first part, structure-based virtual screening (SBVS) methods were employed in the selection and identification of compounds. Based on the in silico design, twenty compounds were selected and experimentally evaluated in the second part using the isothermal titration calorimetry (ITC) technique. Out of these, eleven compounds inhibited the T. cruzi GAPDH, resulting in high hit rates (>= 20 %). The new selected inhibitors display excellent ligand efficiency (LE), as well as some selectivity for the parasite enzyme. The inhibitors assay against the trypomastigote form of T. cruzi was used to identify two compounds able to inhibit this infective form, and one showed to be a strong amastigote parasite inhibitor, also disclosing low cytotoxicity profile. The best two classes of GAPDH and parasite inhibitors were selected for the establishment of a quantitative structure-activity relationship (QSAR). 2D QSAR (HQSAR) studies resulted in linear models with high predictive power, amenable for the identification of important structural features in the process of hit-to-lead optimization.

A doença de Chagas é endemia que afeta grande parte da América Latina. Estima-se que de 18 a 20 milhões de pessoas estejam infectadas pelo parasita causador da doença e que ocorram, aproximadamente, 50.000 mortes relacionadas à doença de Chagas por ano, nos 21 países da área endêmica. O arsenal terapêutico atualmente usado contra a parasitose, constituído por apenas dois fármacos, é insuficiente, considerando-se, também, que ambos não apresentam ação adequada na fase crônica da doença. Em razão de se tratar de parasitose que acomete apenas países em desenvolvimento, o interesse na pesquisa de novos antichagásicos é consideravelmente baixo por parte dos países que se destacam na introdução de novos fármacos na terapêutica. Resta aos países envolvidos a procura por novas alternativas quimioterápicas. Face ao exposto e ante à alta atividade, em testes in vitro contra o Trypanosoma cruzi, do derivado hidroximetilado do nitrofural, intermediário de síntese de bases de Mannich, obtido em trabalhos anteriores, o objetivo do presente projeto foi sintetizar bases de Mannich desse derivado com aminoácidos lisina e arginina e o dipeptídio lisina-arginina resultante. Os compostos foram sintetizados utilizando-se métodos clássicos e alternativos e grupos protetores, normalmente utilizados na síntese de peptídios e outros, tentativamente empregados, como os grupos metílico e etílico. Em adição, experimentos foram efetuados com o objetivo de otimizar a síntese do hidroximetilnitrofural. Os derivados sintetizados foram analisados por IV, RMN 1H e 13C, e alguns deles, também, por espectrometria de massas. Eles serão submetidos a testes in vitro em cultura de células infectadas com T. cruzi tão logo os grupos protetores sejam removidos. Além disso, serão efetuados testes de liberação para estudo da respectiva estabilidade. Dessa forma, esperam-se obter subsídios importantes para estudos mais aprofundados do seu mecanismo de ação e da possível mutagenicidade envolvida. / Chaga\'s disease is an endemic disease that aftects most part of Latin Arnerica. About 18 to 20 million people are infected by the parasite and around 50 thousand deaths are related to Chagas\' disease each year, in the 21 countries of endemic areas. The therapeutic armamentarium available against the disease is constituted by only two drugs and is insufficient, considering, also, that the drugs are not efficient in the chronic phase of the disease. As a parasitosis that only occurs in developing countries, the interest in the research for new antichagasic agents is considerably low in countries that are responsible for the introduction of new drugs in the therapeutics. So, the search for new chemotherapeutic alternatives is a task for the involved countries. In view of the situation and taking into account the high activity in in vitro tests against Trypanosoma cruzi showed by nitrofurazone hydroxymethyl derivative, an intermediary of Mannich bases reaction previously synthesized, the objective of this work was to synthesize its Mannich bases. The carriers used were aminoacids lysine and arginine and its dipeptide, lysinearginine. The compounds were synthesized using classic and alternative methods and protecting groups, currently used in peptide synthesis and others temptatively employed, as methyl and ethyl groups. Also, many experiments were performed in order to achieve the optimization of hydroxymethylnitrofurazone synthesis. The derivatives synthesized were analyzed by IR, 1H and 13C NMR, and some also by mass spectrometry. They will be submitted to in vitro tests with cell infected with T. cruzi as soon as the protecting groups are removed. Besides, tests of drug release will be performed to study their stability. We expect to obtain important information toward better comprehension of their mechanism of action and possible mutagenicity involved.