The current work entails the development of a novel high content platform for the measurement of kinetic ligand responses across cell signalling networks at the single-cell level in distinct PBMC subtypes ex vivo. Using automated sample preparation, fluorescent cellular barcoding and flow cytometry the platform is capable of detecting 21, 840 parallel cell signalling responses in each PBMC sample. We apply this platform to characterize the effects of neuropsychiatric treatments and CNS ligands on the T cell signalling repertoire. We apply it to define cell signalling network abnormalities in PBMCs from drug-naïve first-onset schizophrenia patients (n=12) relative to healthy controls (n=12) which are subsequently normalized in PBMCs from the same patients (n=10) after a six week course of clinical treatment with the atypical antipsychotic olanzapine. We then validate the abnormal cell signalling responses in PBMCs from an independent cohort of drug-naïve first-onset schizophrenia patients (n=25) relative to controls (n=25) and investigate the specificity of the abnormal PBMC responses in schizophrenia as compared to major depression (n=25), bipolar disorder (n=25) and autism spectrum disorder (n=25). Subsequently we conduct a phenotypic drug screen using the US Food and Drug Administration (FDA) approved compound library, in addition to experimental neuropsychiatric drug candidates and nutraceuticals, to identify compounds which selectively normalize the schizophrenia-associated cell signalling response. Finally these candidate compounds are characterized using structure-activity relationships to reveal specific chemical moieties implicated in the putative therapeutic effect.

Dynamic combinatorial chemistry (DCC) is a powerful tool to identify new ligands for biological targets. In the technique, library synthesis and hit identification are neatly combined into a single step. A labile functionality between fragments allows the biological target to self-select binders from a dynamic combinatorial library (DCL) of interconverting building blocks. The scope of suitable reversible reactions that proceed under thermodynamic control in physiological conditions has been gradually expanded over the last decades, however DCC has thus far failed to gain traction as a technique appropriate for drug discovery in the pharmaceutical industry. The constraints placed on library size by validated analytical techniques, and the effort-intensive reality of this academically elegant concept have not allowed DCC to develop into a broad-platform technique to compete with the high-throughput screening campaigns favoured by medicinal chemists. This thesis seeks to develop DCL analysis techniques, in an effort to increase the library size and accelerate the analysis of DCC experiments. Using a 19F-labelled core scaffold, we constructed a DCL that could be monitored non-invasively by 19F NMR. Building on NMR techniques developed by fragment screening and non-biological DCC campaigns, the method was developed to circumvent the undesired equilibrium-perturbing side effects arising from sample-consuming analytical methods. The N-acylhydrazone (NAH) DCL equilibrated rapidly at pH 6.2 using 4-amino-L-phenylalanine (4-APA) as a novel, physiologically benign, nucleophilic catalyst. The DCL was designed to target b-ketoacyl-ACP synthase III (FabH), an essential bacterial enzyme and antibiotic target. From the 5-membered DCL, a single combination was identified as a privileged structure by our 19F NMR method. The result correlated well with an in vitro assay, validating 19F NMR as a tool for DCL screening. During the 19F NMR study we identified an established antimicrobial compound, 4,5- dichloro-1,2-dithiole-3-one (HR45), to have potential as a core scaffold from which to develop future DCLs targeting FabH. Despite the potentially tractable chemistry of HR45 for DCC, lack of knowledge around the inhibitory mechanism of the compound prevented us from proceeding. Thus, we used mass spectrometry, NMR and molecular modelling to show that HR45 acts by forming a covalent adduct with S. aureus FabH. The 5-chloro substituent directs attack from the nucleophilic thiol side chain of the essential active site cysteine-112 residue via a Michael-type addition elimination mechanism. Although interesting, this mechanism disfavoured the use of HR45 as a core scaffold for NAH exchange in a DCC campaign. Electrospray ionisation mass spectrometry (ESI-MS) is a powerful technique that allows for larger DCLs by eliminating the size-limitations imposed by the need for spectral or chromatographic resolution of DCL members. We developed a 4-APAcatalysed NAH library targeting the pyridoxal 5’-phosphate (PLP) dependent enzyme 7,8-diaminopelargonic acid synthase (BioA), an essential enzyme in the biotin biosynthesis pathway. We exploited the aldehyde moiety of PLP to form an NAH DCL with a panel of hydrazides, and used the BioA isozymes from M. tuberculosis (Mtb) and E. coli to template the library. A combination of buffer exchange and denaturing ESI-MS allowed us to conduct a DCC experiment with a 29-member DCL. Hits from the DCC experiment correlated well with differential scanning fluorimetry (DSF) results. Of these hits, 5 compounds were selected for further study. In vivo activity was displayed by 2 compounds against E. coli and the ESKAPE pathogen A. baumannii. The identification of compounds with antibacterial activity from a DCL further validates ESI-MS as a platform technology for drug discovery.

In 2015, an estimated 46.8 million people were living with dementia, a number predicted to increase to 74.7 million by 2030 and 131.5 million by 2050. Whilst there are numerous causes for the development of dementia, Alzheimer's disease is by far the most common, accounting for approximately 50-70% of all cases. Current therapeutic agents against Alzheimer's disease are palliative in nature, managing symptoms without addressing the underlying cause and thus disease progression and patient death remain a certainty. Whilst the main underlying cause for the development of Alzheimer's disease was originally thought to be an abnormal deposition of insoluble amyloid-β peptide derived plaques within the brain, the failure of several high-profile therapeutic agents, which were shown to reduce the plaque burden without improving cognition, has recently prompted a shift in focus to soluble oligomeric forms of amyloid-β peptide. Such soluble oligomers have been shown to be toxic in their own right and to precede plaque deposition. Soluble amyloid-β oligomers have been identified in various subcellular compartments, including the mitochondria, where they form a complex with the 17β-HSD10 enzyme resulting in cytotoxicity. Interestingly, hallmarks of this toxicity have been shown to be dependent on the catalytic activity of the 17β-HSD10 enzyme, suggesting two therapeutic approaches may hold merit in treating Alzheimer's disease: disrupting the interaction between the 17β-HSD10 enzyme and amyloid-β peptide, or directly inhibiting the catalytic activity of the 17β-HSD10 enzyme. In 2006, Frentizole was identified as a small molecule capable of disrupting the 17β-HSD10/amyloid interaction. The work described herein details the generation of a robust screening assay allowing the catalytic activity of the 17β-HSD10 enzyme to be measured in vitro. This assay was subsequently employed for small molecule screening using two methodologies; first in a targeted approach using compounds derived from the Frentizole core scaffold, and second in an explorative manner using a diverse library of compounds supplied by the National Cancer Institute. As a result, a range of novel small molecule inhibitors of the 17β-HSD10 enzyme have been identified and the most promising characterised in terms of potency and mechanism of action. De-selection assays were developed to allow the efficient triage of hit compounds and work was begun on a cellular based assay which would allow the ability of compounds of interest to reverse a disease relevant phenotype to be assessed in a cellular environment. As such, we now have a number of hit compounds which will form the basis for the generation of subsequent series of derivatives with improved potency and specificity, as well as the robust assays required to measure such criteria, potentially leading to the generation of novel therapeutic agents against Alzheimer's disease.

Filariasis and onchocerciasis are debilitating diseases affecting 120 million people globally. The massive socio-economic impact of these diseases energized the international community to declare a goal of eliminating filariasis 2020. This resulted in a dramatic increase in the efforts to eliminate filariasis and onchocerciasis, employing a strategy of mass drug administration (MDA). However, these programs rely upon the small arsenal of drugs. This leaves these programs vulnerable to failure in the face of developing resistance and local intolerance to the current drug regimens. Thus, new drugs against these infections are critically needed. A homologue of the ecdysone receptor (EcR), a master regulator of development in insects, has been identified in B. malayi. The potential of the EcR as a drug target has been underscored by work in the agricultural industry, where insecticides targeting the ecdysone developmental pathway are effective and non-toxic to non-target species. As the EcR is absent in humans, it represents an attractive potential chemotherapeutic target. The first study investigates the hypothesis that the ecdysone receptor controls the embryogenesis and molting in the filarial parasite. In-vitro embryogram and in-vivo phenotypic studies were conducted to delineate the effect of 20-hydroxyecdysone on the Brugia malayi parasites. The results suggest that the hormone accelerates embryogenesis and causes precocious molts, resulting in the death of the parasite. Further, transcriptomic and proteomic analysis of the ecdysone treated worms provided evidence that the up-regulated genes participate in embryogenesis. Based upon the validation of the ecdysone receptor as a potential drug target, subsequent studies focused on the development of a drug discovery model to screen for agonists and antagonists of the B. malayi ecdysone receptor. A stable cell line was created to aid the high throughput screening to rapidly identity agonist and antagonist compounds. A total of 7 agonists and 2 antagonists were identified. A homology model of the BmEcR ligand-binding domain was created as an alternate method for virtual screening of small molecules as well as to study the ligand-receptor interactions. The hits identified with the assay were docked in the active site of the BmEcR homology model providing an excellent correspondence of data between the molecular assay and the virtual screening method.

Due to the inherent problems with their structural elucidation in the laboratory, the computational prediction of membrane protein structure is an essential step toward understanding the function of these leading targets for drug discovery. In this work, the development of a genetic algorithm technique is described that is able to generate predictive 3D structures of membrane proteins in an ab initio fashion that possess high stability and similarity to the native structure. This is accomplished through optimisation of the distances between TM regions and the end-on rotation of each TM helix. The starting point for the genetic algorithm is from the model of general TM region arrangement predicted using the TMRelate program. From these approximate starting coordinates, the TMBuilder program is used to generate the helical backbone 3D coordinates. The amino acid side chains are constructed using the MaxSprout algorithm. The genetic algorithm is designed to represent a TM protein structure by encoding each alpha carbon atom starting position, the starting atom of the initial residue of each helix, and operates by manipulating these starting positions. To evaluate each predicted structure, the SwissPDBViewer software (incorporating the GROMOS force field software) is employed to calculate the free potential energy. For the first time, a GA has been successfully applied to the problem of predicting membrane protein structure. Comparison between newly predicted structures (tests) and the native structure (control) indicate that the developed GA approach represents an efficient and fast method for refinement of predicted TM protein structures. Further enhancement of the performance of the GA allows the TMGA system to generate predictive structures with comparable energetic stability and reasonable structural similarity to the native structure.

Over three decades ago, W.H.O. declared infertility as a public health issue; due to its impact on millions of people worldwide. While cases of infertility could be multifactorial (affecting both male and female), 50% of cases are due to male factor infertility and this is mostly characterised by reduced sperm motility (asthenozoospermia). Assisted Reproduction Technology (ART) is the only treatment option available for this condition. Over 20 years ago, non-selective phosphodiesterase inhibitors (PDEi), such as pentoxifylline, were shown to enhance motility of human spermatozoa; however, contradictory results and stimulation of premature acrosome reaction has precluded their clinical use. Advancement in our knowledge have now made it clear that human sperm express several different PDEs and these are compartmentalised at different regions of the cells. By using type-specific phosphodiesterase inhibitors, differential modulation of sperm motility can be achieved without affecting other sperm function such as acrosome reaction. Additionally, by enhancing sperm function through PDE inhibition, there is a possibility of increasing IVF rates. The objective of this thesis is to: (1) examine the effect of phosphodiesterase inhibitors on spermatozoa in order to identify compounds that have clinically relevant enhancement of human sperm motility; (2) identify the signalling pathway(s) involved in the motility enhancing effects of identified compounds by targeting the modulator and mediator of cyclic nucleotides; (3) develop an animal IVF model to assess effects of Ibudilast on fertilization; and (4) optimise high performance liquid chromatography (HPLC) techniques for routine detection of cyclic nucleotides in sperm cells. A two phase drug screening approach was used to systematically and comprehensively screen series of compounds in order to identify those that have clinically relevant enhancement of human sperm motility. In phase 1, 6 compounds (out of 43 compounds) were found to have strong effects on poor motility samples, with magnitude of response ≥60% increase in percentage total motility. Additionally, these compounds significantly enhanced sperm penetration into cervical mucus substitute (p≤0.05), and they did not affect sperm acrosomal integrity nor cause externalisation of phosphatidylserine (p=0.6 respectively). 63% of IVF samples treated with compounds #26, #37 and #38 had significant increase in percentage total motility. For ICSI samples, compounds #26, #37 and #38 were the most effective. In respect to total motility, 88%, 81% and 79% of samples treated with these compounds showed significant increases in total motility, and 94%, 93% and 81% of samples showed significant increases in percentage of progressive cells, respectively. Analysis of the signalling pathways, using PKA, sGC and PKG inhibitors, showed that chosen PDE inhibitors were working predominantly through PKA signalling pathways. Additionally, this study revealed that this pathway is needed for the maintenance of basal progressive motility and hyperactivation in human sperm. Animal IVF studies showed that addition of Ibudilast (compound #26) during sperm-oocyte incubation leads to higher IVF rates. Lastly, this study used an HPLC system to detect cAMP in boar sperm. This was done to explore if HPLC system can be used for high throughput detection of cyclic nucleotides in mammalian sperm.

Malaria is a parasitic disease that is caused by the plasmodium parasite. Plasmodium infection has affected man for thousands of years. With advances in drug discovery over the past century, malaria has evolved to possess resistance to most mainline therapeutics. This war of drug discovery vs plasmodium evolution continues to be fought to this very day, with attempts to eradicate malaria worldwide. Frontline treatments such as chloroquine, artemisinin, and atovaquone/proguanil have all seen parasitic resistance in strains of P. vivax as well as P. falciparum. While plasmodium possesses resistance to most classes of anti-malarials, the 8-aminoquinoline (8-AQ) class has seen minimal resistance development. 8-AQs have been shown to be effective against erythrocytic and exo-erythrocytic forms of plasmodium, and are often given in combination with a blood schizonticide such as chloroquine or artemisinin. These combinations clear all forms of plasmodium infection. With 8-AQs unique set of anti-malarial properties and the advent of increased drug resistance to other drugs, much research is being done to understand 8-AQs mechanism of action and toxicity. 8-AQ use is limited due to inducing extreme hemolytic anemia in those with glucose-6-phosphate dehydrogenase (G6PD) deficiency. Primaquine is the only 8-AQ molecule available on the market with tafenoquine, an analog primaquine, currently in phase III clinical trials. It is believed that if the mechanism of action and toxicity of the 8-AQs are understood, then we can create new generation anti-malarials that will maintain the unique action of 8-AQs while reducing their toxicity. Studies have shown that 8-AQ mechanism of action has been attributed to the generation of unstable metabolites that induce ROS production in the parasite, as well as mitochondrial swelling. While there is some evidence suggesting molecular targets of 8-AQs, the actual target is still unknown. When 8-AQs is given in combination with chloroquine, a synergistic effect is observed. While chloroquine has no activity against liver stages, it still somehow potentiates primaquine’s activity in those stages. This mechanism of synergy in liver stages is not well understood, and its understanding can give us increased understanding of basic plasmodium biology in the liver. Additionally, more information about the mechanisms of action of both chloroquine and primaquine could be elucidated. Tagging drugs with fluorescent probes is a technique that can give much information about the drug’s pharmacological activity in vitro, and sometimes in vivo as well. Such an approach has been used for various disease states such as HIV and cancer. Malaria is no exception; fluorescent probes of artemisinin and chloroquine have been used to examine resistance mechanisms to both molecules. In addition to 8-AQs, there are other older antimalarials that have received attention recently due to increases in resistance. Menoctone, a hydroxynapthoquinone that subsequently lead to the discovery of atovaquone, has recently gained increased attention because of its similarities to atovaquone. Research surrounding menoctone was abandoned due to the discovery of more efficacious compounds. Similar to 8-AQs, understanding the mechanisms of action and resistance to menoctone could give us much more information about plasmodium responses to this class of compounds. This understanding could potentially lead to the discovery of novel therapeutics. To understand mechanisms of action and synergy of 8-AQs, we report the creation of novel fluorescent probes of the 8-AQ molecules primaquine and tafenoquine. The organic synthesis was designed and characterization was confirmed by NMR and high resolution mass spectra, and the fluorescent properties were examined using absorbance and steady-state emission experiments. We found that the anti-malarial, anti-leishmaniasis, and cytotoxic properties of these novel probes were similar to the parent compounds. These probes localized in the cytoplasm of infected parasites in vitro. We also attempted to view their localization in liver stage infection, and investigated the synergistic combination of 8-AQs with chloroquine and quinine. Menoctone resistance was induced in vivo to determine mechanisms of resistance. Cross resistance to atovaquone was observed, and the mutation responsible for resistance was also found.

Antibiotic resistance has been a developing problem for mankind in recent decades and multi-drug resistant bacteria are now encountered that are resistant to all treatment options available. In 2014, the World Health Organization announced that this problem is driving us towards a “post-antibiotic era” that will change the face of modern medicine as we know it. If lack of novel antibiotic development and FDA approval continues, by the year 2050, 10 million people will die each year to an antimicrobial resistant bacterial infection. With lack of pharmaceutical industry involvement in developing novel antibiotics, the responsibility now lies within the academic institutions to identify potential novel therapeutics to fuel the antibiotic drug discovery pipeline. Combinatorial chemistry is one technique used to expedite the discovery process by assessing a large chemical space in a relatively short time when compared to traditional screening approaches. Combinatorial libraries can be screened using multiple approaches and has shown successful application towards many disease states. We initially discovered broad spectrum antibacterial bis-cyclic guanidines using combinatorial libraries and expanded on the knowledge of the physiochemical attributes necessary to inhibit Gram negative bacterial pathogens. Following this success, we continued to assess the combinatorial libraries for adjunctive therapeutics that potentiate the activity of obsolete clinical antibiotics. The polyamine efflux pump inhibitors discovered in this subsequent study prove the benefits of using the large chemical space provided in the combinatorial libraries to identify a variety of therapeutics. Our studies always begin with identifying an active compound and active compounds undergo hit-to-lead optimization. This optimization studies are of utmost importance in developing a novel antibacterial agent for therapeutic applications. Our medicinal chemistry work described here is proof of the success of careful structure activity analyses to optimize a hit scaffold to create a more effective antibacterial agent. Overall, our work described here reveals the potential role of academic institutions in fending off the impending “post-antibiotic era”.

Abstract INVESTEGATION OF ANTICOAGULATION PROPERTIES OF SULFATED GLYCOSAMINOGLYCAN MIMETICS By Elsamani Ismail Abdelfadiel, MS A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science at Virginia Commonwealth University Virginia Commonwealth University, 2017. Supervisor: Umesh R Desai Professor, Department of Medicinal Chemistry The existence of thrombosis in numerous pathophysiological situations formed a vast necessity for anticoagulation therapy. Thrombin and factor Xa are the only two factors of the entire coagulation cascade that have been major targets for regulation of clotting via the direct and indirect mechanism of inhibition. Our recent discovery of sulfated non-saccharide glycosaminoglycan mimetics, especially G2.2, that demonstrates highly selective cancer stem-like cells (CSCs) inhibition activity. G2.2 inhibited the growth of CSCs from multiple cancer cell lines. To evaluate its in vivo anticoagulation effect, we asked a contract research organization (CRO) to produce 20 g of material, labelled as G2.2Y. Evaluation of G2.2C in HT-29 xenograft mouse model showed a significant reduction in tumor volume and CSC markers, but unexpected bleeding consequences in some animals were observed. Also in a tail bleeding experiment, G2.2Y showed a significant enhancement in bleeding volume. Comparable studies with G2.2 synthesized in our laboratory had shown no bleeding effects. To investigate the difference between the two G2.2 samples (G2.2W (white) and G2.2Y (Yellow) that were performed using UPLC-MS characterization, we were able to determine that the G2.2Y sample was an 85:15 blend of two compounds. Elemental, NMR and MS data revealed that G2.2W was fully sulfated flavonoid derivative, as expected, but G2.2Y contained one less sulfate group. We tested both agents for their inhibition of various coagulation factors and revealed that G2.2Y inhibited fXIa nearly 2-fold better in comparison to G2.2W. Furthermore, activated partial thromboplastin time assay (APTT) indicated that G2.2W exhibited almost 3-4-fold less anticoagulant activity compared to G2.2Y. This indicates that the loss of just one sulfate group could induce substantial side effects and lead to a discovery of new anticoagulant agent. Such structure–activity relationship is important to understand if the in vivo metabolism of the agents leads to accumulation of de-sulfated products.

Structure-based drug design and virtual screening are areas of computational medicinal chemistry that use 3D models of target proteins. It is important to develop better methods in this field with the aim of increasing the speed and quality of early stage drug discovery. The first part of this thesis focuses on the application of structure-based drug design in the search for inhibitors for the protein 1-deoxy-D-xylulose-5-phosphate reductoisomerase (DXR), one of the enzymes in the DOXP/MEP synthetic pathway. This pathway is found in many bacteria (such as Mycobacterium tuberculosis) and in the parasite Plasmodium falciparum. In order to evaluate and improve current virtual screening methods, a benchmarking data set was constructed using publically available high-throughput screening data. The exercise highlighted a number of problems with current data sets as well as with the use of publically available high-throughput screening data. We hope this work will help guide further development of well designed benchmarking data sets for virtual screening methods. Conformal prediction is a new method in the computer-aided drug design toolbox that gives the prediction range at a specified level of confidence for each compound. To demonstrate the versatility and applicability of this method we derived models of skin permeability using two different machine learning methods; random forest and support vector machines.