<p>Drug discovery and development has experienced a tremendous growth in the recent</p> <p>years, and methods to accelerate the process are necessary as the demand for effective drugs</p> <p>to treat a wide range of diseases continue to increase. Nevertheless, the majority of conventional</p> <p>techniques are labor-intensive or have relatively low yields. As a result, academia</p> <p>and the pharmaceutical industry are continuously seeking for rapid and efficient methods to</p> <p>accelerate the drug discovery pipeline. Therefore, in order to expedite the drug discovery</p> <p>process, recent developments in physical and artificial intelligence models have been utilized</p> <p>extensively. However, the overarching problem is how to use these cutting-edge advancements</p> <p>in artificial intelligence to enhance drug discovery? Therefore, this dissertation work</p> <p>focused on developing and applying artificial intelligence and physical models to accelerate</p> <p>the drug discovery pipeline at different stages. As the first study reported in the dissertation,</p> <p>the potential to apply graph neural network-based machine learning architectures</p> <p>with the assistance of molecular modeling features to identify plausible drug leads out of</p> <p>structurally similar chemical databases is assessed. Then, the capability of applying molecular</p> <p>modeling methods including molecular docking and molecular dynamics simulations to</p> <p>identify prospective targets and biological pathways for small molecular drugs is discussed</p> <p>and evaluated in the following chapter. Further, the capability of applying state-of-the-art</p> <p>deep learning architectures such as multi-layer perceptron and recurrent neural networks</p> <p>to optimize the formulation development stage has been assessed. Moreover, this dissertation</p> <p>has contributed to assist functionality identification of unknown compounds using</p> <p>simple machine learning based computational frameworks. The developed omics data analysis</p> <p>pipeline is then discussed in order to comprehend the effects of a particular treatment</p> <p>on the proteome and lipidome levels of cells. In conclusion, the potential for developing and</p> <p>utilizing various artificial intelligence-based approaches to accelerate the drug discovery and</p> <p>development process is explored in this research. Thus, these collaborative studies intend</p> <p>to contribute to ongoing acceleration efforts and advancements in the drug discovery and</p> <p>development field.</p>

Chronic pain affects 20 % of the global population, causes suffering, is difficult to treat, and constitutes a large economic burden for society. So far, the characterization of molecular mechanisms of chronic pain-like behaviors in animal models has not translated into effective treatments. In this thesis, consisting of five studies, pain patient biofluids were analyzed with modern proteomic methods to identify biomarker candidates that can be used to improve our understanding of the pathophysiology chronic pain and lead to more effective treatments. Paper I is a proof of concept study, where a multiplex solid phase-proximity ligation assay (SP-PLA) was applied to cerebrospinal fluid (CSF) for the first time. CSF reference protein levels and four biomarker candidates for ALS were presented. The investigated proteins were not altered by spinal cord stimulation (SCS) treatment for neuropathic pain. In Paper II, patient CSF was explored by dimethyl and label-free mass spectrometric (MS) proteomic methods. Twelve proteins, known for their roles in neuroprotection, nociceptive signaling, immune regulation, and synaptic plasticity, were identified to be associated with SCS treatment of neuropathic pain. In Paper III, proximity extension assay (PEA) was used to analyze levels of 92 proteins in serum from patients one year after painful disc herniation. Patients with residual pain had significantly higher serum levels of 41 inflammatory proteins. In Paper IV, levels of 55 proteins were analyzed by a 100-plex antibody suspension bead array (ASBA) in CSF samples from two neuropathic pain patient cohorts, one cohort of fibromyalgia patients and two control cohorts. CSF protein profiles consisting of levels of apolipoprotein C1, ectonucleotide pyrophosphatase/phosphodiesterase family member 2, angiotensinogen, prostaglandin-H2 D-isomerase, neurexin-1, superoxide dismutases 1 and 3 were found to be associated with neuropathic pain and fibromyalgia. In Paper V, higher CSF levels of five chemokines and LAPTGF-beta-1were detected in two patient cohorts with neuropathic pain compared with healthy controls. In conclusion, we demonstrate that combining MS proteomic and multiplex antibody-based methods for analysis of patient biofluid samples is a viable approach for discovery of biomarker candidates for the pathophysiology and treatment of chronic pain. Several biomarker candidates possibly reflecting systemic inflammation, lipid metabolism, and neuroinflammation in different pain conditions were identified for further investigation. / Uppsala Berzelii Technology Centre for Neurodiagnostics

Dans le monde entier, les infections virales causent des problèmes de santé majeurs et récurrents, engendrant de sérieux problèmes socio-économiques. Notamment, les virus de la famille Flaviviridae qui représentent un fardeau considérable sur la santé mondiale et font partie des domaines prioritaires de la virologie médicale selon le rapport 2016 du ‘Global Virus Network’. Bien que le traitement actuel contre le virus de l’hépatite C (VHC) ait un taux de guérison dépassant 98%, d’autres comme le virus de la dengue (DENV) et le virus zika (ZIKV) n’ont pas encore de traitement spécifique autorisé. En prenant avantage de la grande expertise de notre laboratoire dans l’étude du VHC, nous avons utilisé des données d’une étude de biologie des systèmes visant à identifier l’interactome des différentes protéines virales. Les techniques utilisées ont combiné l’immunoprécipitation des protéines virales suivie de l’identification des protéines interacteurs humaines par spectrométrie de masse. Des études de génomique fonctionnelle par ARN interférent (ARNi) ont permis d’étudier l’effet de la diminution de l’expression des protéines identifiées sur la réplication du VHC. Cette étude a conduit à la découverte de l’interactant spécifique 17-bêta-hydroxystéroïde déshydrogénase de type 12 (HSD17B12 ou DHB12) de la protéine virale Core comme facteur cellulaire requis à la réplication du VHC. HSD17B12 est une enzyme cellulaire dont l’activité catalytique est requise pour l’élongation des acides gras à très longue chaîne (VLCFA) lors de la deuxième des quatre réactions du cycle d’élongation. Dans cette étude, nous avons déterminé que les cycles de réplication du VHC, ZIKV et DENV dépendent de l’expression et de l’activité métabolique du facteur cellulaire HSD17B12. Ainsi, nous avons étudié les effets de l’inhibition de l’expression génique par ARNi et de façon pharmacologique sur la réplication de plusieurs flavivirus dans une approche antivirale à large spectre. Nous avons démontré que le silençage de HSD17B12 diminue significativement la réplication virale, l’expression des protéines virales et la production de particules infectieuses de cellules Huh7.5 infectées par la souche JFH1 du VHC. L'analyse de la localisation cellulaire de HSD17B12 dans des ii cellules infectées suggère une colocalisation avec l'ARN double brin (ARNdb) aux sites de réplication virale, ainsi qu’avec la protéine Core (et les gouttelettes lipidiques) aux des sites d’assemblage du virus. Nous avons également observé que le silençage de HSD17B12 réduit considérablement le nombre et la taille des gouttelettes lipidiques. En accord avec ces données, la diminution de l’expression de HSD17B12 par ARNi réduit significativement l’acide oléique et les espèces lipidiques telles que triglycérides et phosphatidyl-éthanolamine dans l'extrait cellulaire total. Ces travaux suggèrent une contribution de la capacité métabolique de HSD17B12 lors de la réplication du VHC. De même, nous avons démontré que le silençage de HSD17B12 réduit significativement les particules infectieuses de cellules infectées par DENV et ZIKV. Ces études supportent le rôle de HSD17B12 dans l’efficacité des processus de la réplication de l'ARN viral et de l’assemblage de particules virales. De plus, l'inhibiteur spécifique de HSD17B12, INH-12, réduit la réplication du VHC à des concentrations pour lesquelles aucune cytotoxicité notable n'est observée. Le traitement avec 20 μM d'INH-12 réduit jusqu'à 1,000 fois les particules infectieuses produite par des cellules Huh-7.5 infectées par DENV et ZIKV lors de plusieurs cycles de réplication, et bloque complètement l'expression des protéines virales. En conclusion, ces travaux ont conduit à une meilleure compréhension du rôle de HSD17B12 lors de la synthèse de VLCFA et de lipides requise à la réplication du VHC, permettant d’explorer l’inhibition de HSD17B12 et de l’élongation d’acides gras à très longue chaîne comme nouvelle approche thérapeutique pour le traitement à large spectre des infections par les virus de la famille Flaviviridae. / Infections with viruses are major recurrent socio-economical and health problems worldwide. These include infections by viruses of the Flaviviridae family, which present a substantial global health burden and are among the priority areas of medical virology according to the Global Virus Network 2016 report. While the current treatment regimens for hepatitis C virus (HCV) infection have cure rates of more than 98%, other important members of Flaviviridae like dengue virus (DENV) and zika virus (ZIKV) have no specific licensed treatments. By taking advantage of the most-studied HCV, which our lab has developed a vast expertise in the last 20 years, we used proteomics data of an HCV interactome study, combining viral protein immunoprecipitation (IP) coupled to tandem mass spectrometry identification (IP-MS/MS) and functional genomics RNAi screening. The study uncovered the 17-beta-hydroxysteroid dehydrogenase type 12 (HSD17B12, also named DHB12), as a specific host interactor of core that promotes HCV replication. HSD17B12 catalytic activity is involved in the synthesis of very-long-chain fatty acids (VLCFA) upon the second step of the elongation cycle. In this study, taking HCV as a virus model, we elucidated the dependency of HCV, dengue virus (DENV) and zika virus (ZIKV) replication on expression and metabolic capacity of the host factor HSD17B12. We investigated the effects of the inhibition of gene expression by RNAi and of its pharmacological enzymatic inhibition on flavivirus replication in a broad-spectrum antiviral approach. We showed that silencing expression of HSD17B12 decreases viral replication, viral proteins and iv infectious particle production of the JFH1 strain of HCV in Huh7.5 cells. The cellular localization analysis of HSD17B12 showed a co-staining with double-stranded RNA (dsRNA) at viral replication sites and with core protein (and lipid droplets) at virus assembly sites. Furthermore, HSD17B12 gene silencing drastically reduced the number and size of lipid droplets. In association, the reduced expression of HSD17B12 by RNAi decreases oleic acid levels and lipids such as triglycerides (TG) and phosphatidylethanolamine (PE) in whole-cell extract. The data suggested the requirement of the metabolic capacity of HSD17B12 for HCV replication. Similarly, we provide evidence that HSD17B12 silencing significantly reduces DENV and ZIKV infectious particles. The studies support a role of HSD17B12 for effective viral RNA replication and particle assembly processes. Moreover, the specific HSD17B12 inhibitor, INH-12, reduces HCV replication at concentrations for which no appreciable cytotoxicity is observed. The treatment of DENV- and ZIKV-infected Huh- 7.5 cells with 20 μM of INH-12 dramatically reduces production of infectious particles by up to 3-log10 in infection assays, and completely block viral protein expression. In conclusion, these studies extends our understanding of the role of HSD17B12 in VLCFA synthesis required for the replication of HCV, allowing to explore the inhibition of HSD17B12 and elongation of VLCFA as a novel therapeutic approach for the treatment of a broad-spectrum of viruses of the Flaviviridae family.

<div>Delayed fracture healing is a major health issue involved with aging. Therefore, strategies to improve the pace of repair and prevent non-union are needed in order to improve patient outcomes and lower healthcare costs. In order to accelerate bone fracture healing noninvasively, we sought to develop a drug delivery system that could safely and effectively be used to deliver therapeutics to the site of a bone fracture. We elected to pursue the promising strategy of using small-molecule drug conjugates that deliver therapeutics to bone in an attempt to increase the efficacy and safety of drugs for treating bone-related diseases.</div><div>This strategy also opened the door for new methods of administering drugs. Traditionally, administering bone anabolic agents to treat bone fractures has relied entirely on local surgical application. However, because it is so invasive, this method’s use and development has been limited. By conjugating bone anabolic agents to bone-homing molecules, bone fracture treatment can be performed through minimally invasive subcutaneous administration. The exposure of raw hydroxyapatite that occurs with a bone fracture allows these high-affinity molecules to chelate the calcium component of hydroxyapatite and localize primarily to the fracture site.</div><div>Many bone-homing molecules (such as bisphosphonates and tetracycline targeting) have been developed to treat osteoporosis. However, many of these molecules have toxicity associated with them. We have found that short oligopeptides of acidic amino acids can localize to bone fractures with high selectivity and with very low toxicity compared to bisphosphonates and tetracyclines.</div><div>We have also demonstrated that these molecules can be used to target peptides of all chemical classes: hydrophobic, neutral, cationic, anionic, short, and long. This ability is particularly useful because many bone anabolics are peptidic in nature. We have found that acidic oligopeptides have better persistence at the site of the fracture than bisphosphonate-targeted therapeutics. This method allows for a systemic administration of bone anabolics to treat bone fractures, which it achieves by accumulating the bone anabolic at the fracture site. It also opens the door for a new way of treating the prevalent afflictions of broken bones and the deaths associated with them.</div><div>We further developed this technology by using it to deliver anabolic peptides derived from growth factors, angiogenic agents, neuropeptides, and extracellular matrix fragments. We found several promising therapeutics that accelerated the healing of bone fractures by improving the mineralization of the callus and improving the overall strength. We optimized the performance of these molecules by improving their stability, targeting ligands, linkers, dose, and dosing frequency.</div><div>We also found that these therapeutics could be used to accelerate bone fracture repair even in the presence of severe comorbidities (such as diabetes and osteoporosis) that typically slow the repair process. We found that, unlike the currently approved therapeutic for fracture healing (BMP2), our therapeutics improved functionality and reduced pain in addition to strengthening the bone. These optimized targeted bone anabolics were not only effective at healing bone fractures but they also demonstrated that they could be used to speed up spinal fusion. Additionally, we demonstrated that acidic oligopeptides have potential to be used to treat other bone diseases with damaged bone.</div><div>With these targeted therapeutics, we no longer have to limit bone fracture healing to casts or invasive surgeries. Rather, we can apply these promising therapeutics that can be administered non-invasively to augment existing orthopedic practices. As these therapeutics move into clinical development, we anticipate that they will be able to reduce the immobilization time that is the source of so many of the deadly complications associated with bone fracture healing, particularly in the elderly.</div>