Exploration of peroxisome proliferator-activated receptor-gamma (PPAR") as a drug target holds applications for treating a wide variety of chronic inflammation-related diseases. Type 2 diabetes (T2D), which is a metabolic disease influenced by chronic inflammation, is quickly reaching epidemic proportions. Although some treatments are available to control T2D, more efficacious compounds with fewer side effects are in great demand. Drugs targeting PPAR" typically are compounds that function as agonists toward this receptor, which means they bind to and activate the protein. Identifying compounds that bind to PPAR" (i.e. binders) using computational docking methods has proven difficult given the large binding cavity of the protein, which yields a large target area and variations in ligand positions within the binding site. We applied a combined computational and experimental concept for characterizing PPAR" and identifying binders. The goal was to establish a time- and cost-effective way to screen a large, diverse compound database potentially containing natural and synthetic compounds for PPAR" agonists that are more efficacious and safer than currently available T2D treatments. The computational molecular modeling methods used include molecular docking, molecular dynamics, steered molecular dynamics, and structure- and ligand-based pharmacophore modeling. Potential binders identified in the computational component funnel into wet-lab experiments to confirm binding, assess activation, and test preclinical efficacy in a mouse model for T2D and other chronic inflammation diseases. The initial process used provided "-eleostearic acid as a compound that ameliorates inflammatory bowel disease in a pre-clinical trial. Incorporating pharmacophore analyses and binding interaction information improved the method for use with a diverse ligand database of thousands of compounds. The adjusted methods showed enrichment for full agonist binder identification. Identifying lead compounds using our method would be an efficient means of addressing the need for alternative T2D treatments. / Ph. D.

This thesis describes the development and refinement of a number of techniques for molecular docking and ligand database screening, as well as the application of these techniques to predict the structures of several protein-ligand complexes and to discover novel ligands of an important receptor protein. Global energy optimisation by Monte-Carlo minimisation in internal co-ordinates was used to predict bound conformations of eight protein-ligand complexes. Experimental X-ray crystallography structures became available after the predictions were made. Comparison with the X-ray structures showed that the docking procedure placed 30 to 70% of the ligand molecule correctly within 1.5A from the native structure. The discrimination potential for identification of high-affinity ligands was derived and optimised using a large set of available protein-ligand complex structures. A fast boundary-element solvation electrostatic calculation algorithm was implemented to evaluate the solvation component of the discrimination potential. An accelerated docking procedure utilising pre-calculated grid potentials was developed and tested. For 23 receptors and 63 ligands extracted from X-ray structures, the docking and discrimination protocol was capable of correct identification of the majority of native receptor-ligand couples. 51 complexes with known structures were predicted. 35 predictions were within 3A from the native structure, giving correct overall positioning of the ligand, and 26 were within 2A, reproducing a detailed picture of the receptor-ligand interaction. Docking and ligand discrimination potential evaluation was applied to screen the database of more than 150000 commercially available compounds for binding to the fibroblast growth factor receptor tyrosine kinase, the protein implicated in several pathological cell growth aberrations. As expected, a number of compounds selected by the screening protocol turned out to be known inhibitors of the tyrosine kinases. 49 putative novel ligands identified by the screening protocol were experimentally tested and five compounds have shown inhibition of phosphorylation activity of the kinase. These compounds can be used as leads for further drug development.

Virtual Sreening (VS) is an in silico technique for drug discovery. An overview of VS methods is given and is seen to be approachable from two sides: structure based and ligand based. Structure based virtual screening uses explicit knowledge of the target receptor to suggest candidate receptor-ligand complexes. Ligand based virtual screening can infer required characteristics of binders from known ligands. A consideration for all virtual screening techniques is the amount of computing time required to arrive at a solution. For this reason, techniques of high performance computing have been applied to both the structural and ligand based approaches. A proven structure based virtual screening code LIDAEUS (Ligand Discovery At Edinburgh University) has been ported and parallelised to a massively parallel computing platform, the University of Edinburgh’s IBM Bluegene/l, consisting of 2,048 processor cores. A challenge in achieving scaling to such a large number of processors required implementation of a minimal communication parallel sort algorithm. Parallel efficiencies achieved within this parallelisation exceeded 99%, confirming that a near optimum strategy has been followed and capacity for running the code on a greater number of processors exists. This implementation of the program has been successfully used with a number of protein targets. The development of a new ligand based virtual screening code has been completed. The program UFSRAT (Ultra Fast Shape Recognition with Atom Types) takes the features of known binders and suggests molecules which will be able to make similar interactions. This similarity method is both fast (1 million molecules per hour per processor) and independent of input orientation. Along with UFSRAT, some other methods (VolRAT and UFSRGraph) based on UFSRAT have been developed, addressing different approaches to ligand based virtual screening. UFSRAT as an approach to discovering novel protein-ligand complexes has been validated with the discovery of a number of inhibitors for 11β-Hydroxysteroid Dehydrogenase type 1 and FK binding protein 12.

Computational methods have become an integral part of drug development and can help bring new and better drugs to the market faster. The process of predicting the biological activity of large compound collections is known as virtual screening, and has been instrumental in the development of several drugs today in the market. Computational methods can also be used to elucidate the energies associated with chemical reactivity and predict how to improve a synthetic protocol. These two applications of computational medicinal chemistry is the focus of this thesis. In the first part of this work, quantum mechanics has been used to probe the energy surface of palladium(II)-catalyzed decarboxylative reactions in order to gain a better understating of these systems (paper I-III). These studies have mapped the reaction pathways and been able to make accurate predictions that were verified experimentally. The other focus of this work has been to develop virtual screening methodology. Our first study in the area (paper IV) investigated if the results from several virtual screening methods could be combined using data fusion techniques in order to get a more consistent result and better performance. The study showed that the results obtained from data fusion were more consistent than the results from any single method. The data fusion methods also for several target had a better performance than any of the included single methods. Next, we developed a dataset suitable for evaluating the performance of virtual screening methods when applied to large compound collection as a replacement or complement for high throughput screening (paper V). This is the first benchmark dataset of its kind. Finally, a method for using computationally derived reaction coordinates as basis for virtual screening was developed. The aim was to find inhibitors that resemble key steps in the mechanism (paper VI). This initial proof of concept study managed to locate several known and one previously not reported reaction mimetics against insulin regulated amino peptidase.

Drug discovery has witnessed an increase in the application of in silico methods to complement existing in vitro and in vivo experiments, in an attempt to 'fail fast' and reduce the high attrition rates of clinical phases. Computer algorithms have been successfully employed for many tasks including biological target selection, hit identification, lead optimization, binding affinity determination, ADME and toxicity prediction, side-effect prediction, drug repurposing, and, in general, to direct experimental work. This thesis describes a multifaceted approach to virtual screening, to computationally identify small-molecule inhibitors against a biological target of interest. Conformer generation is a critical step in all virtual screening methods that make use of atomic 3D data. We therefore analysed the ability of computational tools to reproduce high quality, experimentally resolved conformations of organic small-molecules. We selected the best performing method (RDKit), and developed a protocol that generates a non-redundant conformer ensemble which tends to contain low-energy structures close to those experimentally observed. We then outline the steps we took to build a multi-million, small-molecule database (including molecule standardization and efficient exact, substructure and similarity searching capabilities), for use in our virtual screening experiments. We generated conformers and descriptors for the molecules in the database. We tagged a subset of the database as `drug-like' and clustered this to provide a reduced, diverse set of molecules for use in more computationally-intensive virtual screening protocols. We next describe a novel virtual screening method we developed, called Ligity, that makes use of known protein-ligand holo structures as queries to search the small-molecule database for putative actives. Ligity has been validated against targets from the DUD-E dataset, and has shown, on average, better performance than other 3D methods. We also show that performance improved when we fused the results from multiple input structures. This bodes well for Ligity's future use, especially when considering that protein structure databases such as the Protein Data Bank are growing exponentially every year. Lastly, we describe the fruitful application of structure-based and ligand-based virtual screening methods to Plasmodium falciparum Subtilisin-like Protease 1 (PfSUB1), an important drug target in the human stages of the life-cycle of the malaria parasite. Our ligand-based virtual screening study resulted in the discovery of novel PfSUB1 inhibitors. Further lead optimization of these compounds, to improve binding affinity in the nanomolar range, may promote them as drug candidates. In this thesis we postulate that the accuracy of computational tools in drug discovery may be enhanced to take advantage of the exponential increase of experimental data and the availability of cheaper computational power such as cloud computing.

In all kinds of disease models, many proteins involved in protein-protein interactions (PPIs) are mutated and do not function properly. The important role of PPIs in disease makes the design of small molecule inhibition an interesting proposition. This project looks at mouse double minute 2 (MDM2) and mouse double minute X (MDMX) which binds and inhibits the tumour suppressor protein p53. MDM2 and MDMX are therefore attractive therapeutic targets due to their role in tumour progression. The aim is to identify small molecule dual inhibitors that are able to disrupt MDM2 and MDMX from binding to p53. Both N-terminal MDM2 and MDMX were successfully expressed and purified with high purity and decent yield. These proteins were used to develop Fluoresence Polarization (FP) and Capillary Electrophoresis (CE) assays for small molecule inhibitors screening. This work has successfully developed FP and CE assays for detecting weakly interacting fragments. The CE assay is a novel method for detecting weak fragments for protein-protein interactions, which are a challenging target. Two approaches were employed to identify small molecule inhibitors for MDM2- N/p53 interaction. At first, small molecules were identified using in silico screening and these hits were verified using FP and CE assays. Second, analogue exploration was applied to identify fragments from the small molecule inhibitors discovered from the in silico screening. Diphenylamine and oxindole fragments were identified as the most potent. However, diphenylamine fragment was discovered to aggregate MDM2-N and was ranked as a false positive hit. No protein aggregation was found when incubated with the oxindole fragment. Therefore oxindole can provide a good starting point for the design of higher affinity analogues. Studying the interaction of MDMX has only recently been undertaken. MDMX contains a high homology binding site with MDM2. Hence, developing a dual MDM2/MDMX inhibitor has become an attractive target to focus on. FP and CE assays were developed to screen compounds against MDMX-N. In silico screening against MDM2-N and MDMX-N found several hits. One compound was discovered as a dual binder to MDM2-N and MDMX-N with low μM affinity. This novel hit is potentially a good starting point for the design of higher affinity analogues.

MDM2, murine double minute 2, is a RING type-E3 ligase protein and also an oncogene. MDM2 plays a critical role in determining the steady levels and activity of p53 in cells using two mechanisms. The N-terminal domain of MDM2 binds to the transactivation domain of p53 and inhibits its transcriptional activity. The RING domain of MDM2 plays a role in the ubiquitination (and degradation) of p53. Several proteins are responsible for the ubiquitination mechanism including the ubiquitin-activating enzyme (E1), ubiquitin-conjugating enzyme (E2) and ubiquitin ligase (E3). Since the E2-E3 interaction is essential for ubiquitination, the protein-protein recognition site is a potential drug target. Two different MDM2 RING constructs were expressed and purified: MDM2RING (residues 386-491) and MDM2RING△C (residues 386-478). Both constructs were characterised using dynamic light scattering, size exclusion chromatography, mass spectrometry, NMR and electron microscopy. E3 ligase activity in vitro was also studied. Taken together these results showed that the MDM2RING construct formed a concentration-dependent oligomeric structure. In contrast, the MDM2RING△C construct formed a dimer at all concentrations. Both MDM2RING and MDM2RING △ C retain E3 ligase activity. However, the MDM2RING△C construct is less active. Full length E2 enzyme UbcH5a was also purified. Various biophysical techniques were used to study its interaction with MDM2 as well as with potential small molecule inhibitors as in principle, small molecules which disrupt the interaction between MDM2 and UbcH5a, could prevent/promote ubiquitination of p53. The dimerisation of MDM2 is important for its E3 activity and the C8-binding site potentially provides a second druggable site. In this work, peptide 9, which has the same sequence as the C-terminus of MDMX (an MDM2 homologue) was found to inhibit MDM2 E3 activity. Various biological techniques including NMR, fluorescence anisotropy, and electrospray mass spectrometry were used to investigate the interaction between two inhibitory peptides and MDM2. A major part of project involved virtual screening (VS) to search for small molecules which can affect MDM2-dependent ubiquitination. Three potential targets were considered: (1) the C8-binding site of MDM2; (2) the UbcH5a-binding site of MDM2; and (3) the MDM2-binding site of UbcH5a. Several small molecules were identified using our virtual screening database-mining and docking programs that were shown to affect MDM2-dependent ubiquitination of p53. In terms of understanding the complex biochemical mechanism of MDM2 this work provides two interesting and functionally relevant observations: (i) the MDM2 RING△C construct is a dimer as this would not be expected form the existing studies, and has less E3 ligase activity than MDM2RING; (ii) small molecules that bind MDM2 on the E2 binding site enhanced E3 ligase activity. One model to explain these observations is that binding of small molecule activators family to the RING induces a change in the conformation of the Cterminal tail residues which may enhance E2 binding.

Due to the development of multidrug resistance in Streptococcus pneumoniae, research has begun to define new drug targets for pneumonia therapy. Different research groups have identified a lipoprotein, PsaA that is important for pneumonia virulence. PsaA is a manganese transporter that is required for bacterial virulence and growth. We have employed computer modeling to virtually screen a small-molecule database for inhibition of PsaA function by targeting the metal binding pocket, performing receptor-based virtual screening and molecular docking and scoring to identify potential inhibitors of PsaA function. We have developed an assay for screening compounds, including the use of a PsaA mutant, testing of multiple compounds, and identification of compounds that inhibit Streptococcus pneumoniae growth at concentrations less than 20 μM. We experimentally tested the effect on Mn uptake and their PsaA dependence for 42 compounds, but these experiments suggested that these compounds were affecting bacterial growth by a different mechanism.

Staphyloccocus aureus is an opportunisitic pathogen that causes soft skin and tissue infections (SSTI) such as endocarditis, osteomyelitis and meningitis. In recent years, the re-emergence of antibiotic-resistant S. aureus such as MRSA presents a formidable challenge for infection management worldwide. Amidst this global epidemic of antimicrobial resistance, several research efforts have turned their focus towards exploiting the cell-wall biosynthesis pathway for novel anti-bacterial targets. Recently, the lipoteichoic acid (LTA) biosynthesis pathway has emerged as a potential anti-bacterial target. LTA is an anionic polymer found on the cell envelope of Gram-positive bacteria. It comprises of repeating units of glycerol-phosphate (GroP) and is important for bacterial cell physiology and virulence. For example, it is critically involved in regulating ion homeostasis, cell division, host colonization and immune system invasion. Several reports showed that bacteria lacking LTA are unable to grow. At the same time, they suffer from severe cell division defects and also exhibit aberrant cell morphologies. The key protein involved in the LTA biosynthesis pathway is the Lipoteichoic acid synthase (LtaS). LtaS is located on the cell membrane of Gram-positive bacteria and can be divided into two parts: a transmembrane domain and an extra-cellular domain responsible for its enzymatic activity (annotated eLtaS). Given that LtaS is important for bacterial survival and there are no known eLtaS homologues in eukaryotic cells, this protein is an attractive antibacterial target. In 2013, a small molecule eLtaS inhibitor (termed 1771) was discovered. Although 1771 was able to deplete LTA production, the binding mechanism of 1771 to eLtaS remains unknown. Additionally, 1771 could only prolong the survival of infected mice temporarily because of its in vivo instability. Therefore, the need for finding more potent and metabolically stable inhibitors of eLtaS still remains. Computational-aided drug design (CADD) is a cost-effective and useful approach that has been widely integrated into the drug discovery process. The protein eLtaS lends itself to be a good target for CADD since its crystal structure and a known inhibitor (with limited structure-activity data) is available. In this work, I have targeted eLtaS using CADD methodology followed by prospective validation using various biophysical, biochemical and microbiological assays. My project can broadly be sub-divided into three phases: (a) identification of small molecule binding “hot spots”, (b) optimization of existing inhibitor and (c) discovery of new hits. Through a systematic use of different computational approaches, I modelled a plausible 1771-bound eLtaS complex and used the structural insights to generate new inhibitors against eLtaS. To this end, I discovered EN-19, which is a more potent inhibitor of eLtaS. Additionally, by targeting transient cryptic pockets predicted by Molecular Dynamic simulations, I have discovered a new inhibitor chemotype that seems to exhibit a different mode of action against eLtaS. Taken together, my work presents a computational platform for future drug design against eLtaS and reinforces the notion that targeting eLtaS is a viable strategy to combat Gram-positive infections.

A major area of interest is the identification of proteins that play a role in hormone dependent cancers and in collaboration with the MRC Centre for Reproductive Health we studied the gonadotropin releasing hormone receptor (GnRH-R). Other targets described in the thesis are the SH3 domain of PSD-95 and the protein BLyS. In order to identify potential inhibitory small molecules we have used a variety of computational data base mining approaches as well as using and developing experimental binding assays. It has become increasingly challenging to evaluate the most representative drug like small molecule compounds when using traditional high throughput screening methods. This thesis assesses the use of in silico tools to probe key protein-protein and protein-peptide interactions. These tools provide a means to identify enriched compound datasets which can be purchased and tested in vitro in a time and cost efficient way. The transmembrane protein GnRH-R provides an interesting opportunity to identify small molecules that could inhibit the binding of its peptide ligand GnRH. This is a challenging project as there are few examples in the literature of drug-like molecules that bind to such protein-peptide interfaces. The first step involved receptor modelling using solved crystal structures of homologous proteins. The model was then validated by developing structure activity relationships for established high affinity ligands. We also performed crystallographic and biophysical studies on the native GnRH decapeptide. Two other protein-protein systems were also examined using the same virtual screening and experimental ligand binding methodology. SH3 domains play an important role in cell signalling and we used the PSD-95 protein as our target for study as a crystal structure has been published. As well as identifying potential ligands we characterised structural properties of PSD-95 fusion proteins and also developed the basis for compound assay. The third system studied was B Lymphocyte Stimulator (BLyS) which is a target for treatment of a number of autoimmune diseases. This presented an interesting target for study as the protein binds to multiple receptors depending on its multimeric state. BLyS protein was characterised using electron microscopy and other biophysical techniques.