Search

Alternative Splicing is one of the most important post-transcriptional processes as it facilitates the formation of multiple protein isoforms from the same gene sequence. Despite alternative splicing being highly regulated, anomalous splicing behaviour can be the root cause of disease states such as cancer. To date, research efforts have looked to explore and understand aspects of the spliceosome machinery but elucidating the effect of changes in the gene sequence upon splicing is significantly lacking. To define this critical effect, improved site-specific labelling of RNA is required. This thesis focuses on the synthesis and development of an orthogonal non-natural base pair system for the purpose of improving current RNA labelling strategies. Chapter one introduces alternative splicing, describes the regulatory processes involved and current methodologies for RNA labelling and their limitations. This chapter also introduces non-natural base pairs as a solution to overcome these limitations as well as other applications for non-natural base pairs. Chapter two describes the development of a library of guanosine analogues using a modular step efficient synthetic approach with the future aim of site specifically labelling viral RNA. Therefore this library was tested for antiviral activity and a trio of purine base analogues were tested for pairing fidelity opposite Hirao's complimentary base to his purine analogues, the pyrrole analogue Pa by transcription assay. The antiviral testing was promising as no antiviral activity was found, however, the transcription assay did not show improvement over the Pa-s reference standard. Chapter three examines the utility of a new ribose hydroxyl protection strategy with the aim of achieving selective phosphoramidite synthesis of RNA nucleosides to improve the overall yield of non-natural purine phosphoramidites for solid phase RNA synthesis. Synthesis of a non-natural purine phosphoramidite precursor was achieved utilising the chosen strategy. Chapter Four describes the development of a modular synthetic route for 2'-deoxyadenosine phosphoramidite analogues, their incorporation into DNA by solid-phase DNA synthesis and an investigation of their utility for use as DNA aptamers. Incorporation of non-natural phosphoramidites into a 12mer DNA strand was achieved, however, the thermal stability test did not show improved pairing strength opposite Pa compared to the Ds reference standard.

Alternative splicing is the primary source of protein diversity in eukaryotes. Alternative splicing allows the production of multiple isoforms from the same genetic blueprint (DNA). Over 95% of all eukaryotic genes undergo alternative splicing, and although this facilitates great protein diversity, it can instigate disease when the process goes wrong for example in cancer. The Bcl-2 protein family are essential gatekeeper regulators of apoptosis. One such protein, Bcl-x is of particular interest as a therapeutic target. It has two splicing isoforms; pro apoptotic Bcl-xS and the anti-apoptotic Bcl-xL; the latter of which is upregulated in a variety of cancers. Thus, exogenous regulation of Bcl-x splicing, which biases the pathway towards the pro-apoptotic Bcl-xS isoform could provide a new a novel mechanism for cancer therapy. In 2018, Eperon and co-workers found that the ellipticine analogue GQC-05 induces a splice switch in Bcl-x towards the pro-apoptotic isoform Bcl-xS. This thesis will report the steps taken towards developing a structure activity relationship (SAR) of the interaction between GQC-05 and Bcl-x in an effort to elucidate the splice-switch mode of action. Chapter 1 discusses the biological importance of splicing and its effect on disease progression. In particular, the recent developments highlighting Bcl-x as a desirable target for novel cancer therapies. Chapter 2 describes a modular one-pot synthesis of 7H-pyridocarbazoles through sequential palladium catalysed Buchwald-Hartwig amination and direct arylation. This method was transferrable to the synthesis of carbazole natural products as well as providing a convergent step-efficient route to the known DNA intercalator, Ditercalinium. Chapter 3 explores the mechanism of the direct arylation step through designed control reactions and modelling reactivity through DFT calculations. Finally, Chapter 4 details the application of the methods developed in Chapter 2 to the synthesis of natural products Calothrixin B and Ellipticine. This chapter discusses the limitations of the one-pot synthesis on substrates that are more complex and presents a UV light mediated synthesis as a solution. Finally, the ellipticine analogues synthesised in this thesis were tested in Bcl-x in vitro splicing assays to expand the SAR of the interaction to inform the design of a structural-biological probe.

Idiopathic pulmonary fibrosis (IPF) is a progressive and fatal disease of the lung, characterised by excessive collagen deposition. A median survival time of 2-3 years gives IPF a worse prognosis than many cancers. The two currently approved treatments only slow progression of the disease, while the best treatment - a lung transplant - has long waiting lists, risks associated with organ rejection and is not a viable option for all patients. Researchers at GlaxoSmithKline (GSK) have demonstrated that inhibition of mTOR kinase halts the deposition of collagen. Research is underway to develop an inhaled small molecule inhibitor of mTOR kinase for the treatment of IPF. Chapter I describes the lead optimisation of a novel series of directly-linked sulfone mTOR kinase inhibitors (A, Figure). Compounds were designed and synthesised to explore structure activity relationships (SAR) by modifying the substituents at the 2-,4-, and 6-positions (R1, R2 and R3). The aims of this research were to: a) identify optimal combinations of the three substituents; b) synthesise compounds that met the target property profile (suitable efficacy, affinity and selectivity with no mutagenic risks); and c) investigate a range of novel sulfone moieties. While these aims were achieved, an in vivo study demonstrated that compounds from within this directlylinked sulfone series did not have the required in vivo efficacy. Figure: The two series of mTOR kinase inhibitors discussed in this Thesis. A - the directly-linked sulfone series and B - the carbon-linked sulfone series. Substituents in each of the three vectors coloured in red, green and blue. [Graphic element here] In parallel to the research into the directly-linked sulfone compound, an alternative series of carbon-linked sulfone mTOR kinase inhibitors was explored (B, Figure). As compounds in this series were progressed, larger quantities were needed, requiring a robust synthesis. Chapter II describes the work carried out to overcome the three key challenges in the synthesis: 1) To develop an improved synthesis of the 1-(5-chloro-1H-pyrrolo[3,2-b]pyridin-2-yl)-N-methylmethanamine moiety (E, Scheme, here referred to as the azaindole). 2) To improve the SNAr reaction conditions to install the (S)-3-ethylmorpholine to give compound C (Scheme). 3) To identify a suitable cross-coupling methodology to couple the 2-chloropyridine core (C) and the azaindole (E) to give final compound F. An improved synthesis of the azaindole (E) was designed, employing the Larock indole synthesis and a solvent and base screen gave improved conditions for the SNAr reaction to give compound C. Originally only feasible using Stille chemistry and toxic organostannanes, the bipyridyl cross-coupling reaction was improved by employing a desulfinative cross-coupling reaction (coupling D and E to give F, Scheme). Scheme: Forming the sulfinate and the successful desulfinative cross-coupling reaction. Reagents and Conditions: i) SMOPS (sodium 3-methoxy-3-oxopropane-1-sulfinate), Cu(I)I, DMSO, 110 °C. ii) NaOMe (0.5 M in MeOH), THF, 21 °C. iii) K2CO3, Pd(OAc)2, PCy3, 1,4-dioxane, 150 °C. iv) HCl (4 M in 1,4-dioxane), 1,4-dioxane. [Graphic element here] The scope of this desulfinative cross-coupling reaction was subsequently explored in Chapter III, with a particular focus on the synthesis of bipyridyl compounds. This demonstrated that a range of bipyridyl compounds could be made by this method. Finally, a high-throughput screening platform for the desulfinative cross-coupling reaction was designed and validated giving a rapid method to screen catalysts, bases and solvents and optimise the reaction. This proved particularly valuable to develop conditions for some of the more challenging cross-coupling substrates.

Point of Care devices have helped extend the reach of medical services and improved turnaround times for many traditional laboratory tests. Their development necessitates the development and miniaturisation of novel sensing schemes for rapid monitoring of nanoscale biochemical interactions. Many pathologies such as cardiovascular disease present physical biomarkers in patient samples. By probing at the surface of a thin-film, organic semiconductor, distributed feedback laser a novel biosensing scheme utilizing oligonucleotides has been studied and is reported in this thesis. Using an optical approach, Truxene and BBEHP-PPV DFB lasers are put to the test, assessing their sensing capabilities using laser wavelength shift as the transducing method. Bulk sensitivities of 23 nm.RIU-1 (Tr) and 21 nm.RIU-1 (BBEHP_PPV) are observed. The Tr-DFB lasers demonstrated higher sensitivity in bulk superstrate and surface material adsorption tests. As an alternative to wavelength shift, silver nanopa rticle labelled oligonucleotide targets were used to influence the threshold fluence required for OS-DFB lasers to achieve laser action. Using this method, a successful immobilisation strategy for oligonucleotide probes on the surface of an OS-DFB laser, for the first time to our knowledge, was confirmed. Target AgNPs were detected by SERS with a 20x objective on a Raman microscope (Gloucestershire, UK) equipped with an excitation wavelength of 514.5 nm and comparative concentration dependant effects on OS-DFB laser threshold fluence down to 11.5 pM. The capture of 40-45 nm AgNP surface bound target oligonucleotides via hydrogen base pair bonding to corresponding probes on the OS-DFB laser surface was shown to be selective down to individual nucleotide. Such detection limits are within diagnostic significance and the protocols have scope for modification to suit a multitude of different applications.The novel demonstration of a molecular attachment strategy for oligonucleotide biosensor probes to an OS-DFB laser surface and the detection of AgNPs bound by complementary strand hybridisation using SERS and a comparative threshold fluence detection method are of particular note. The text within explores the characteristics of these sensing regimes and compares and contrasts the biosensing technologies involved.

The alkylation of aromatic small molecules is an essential synthetic process used throughout medicinal chemistry. Key limitations of synthetic methods such as the Friedel-Crafts reaction is its lack of chemoselectivity in complex organic scaffolds and stoichiometric amounts of metallic waste produced by this process. A potential solution to this issue could be the use of methyltransferase (MTase) enzymes as biocatalysts to carry out this process. In nature, Sadenosyl-L-methionine (SAM) dependent MTases carry out such reactions on a variety of substrates ranging from small molecules to DNA and proteins. SAM donates its methyl groupto the target substrate with S-adenosyl-L-homocysteine (SAH) generated as a stoichiometric byproduct. The MTase NovO regio specifically transfers alkyl groups from chemically synthesised SAM analogues to a coumarin scaffold which is a precursor to the naturally-occurring antibiotic novobiocin. Chemical synthesis of SAM analogues has the issue of producing a diastereomeric mixture with epimers generated at the sulfonium centre, and display limited stability. SAMcan be generated biosynthetically from the reaction of 5â€2-deoxy-5â€2-chloroadenosine (ClDA)with L-methionine (L-met), catalysed by an enzyme, SalL, from Salinospora tropica. Thismethod is particularly attractive as ClDA and L-met are cheap and produce only the biologically active (S,S) diastereomer. This thesis is a study of both SalL and NovO and their use in tandem SAM generation/alkylation (Scheme). The crystal structure of SalL has been solved to guide Site-Directed Mutagenesis (SDM). This method has then been used to better understand key residues in substrate binding and enzyme activity. Further to this, rational mutations were made in an attempt to increase space in the binding pocket to allow L-met analogues bearing groups larger than methyl to be accommodated. To this end, a range of L-met analogues were synthesised. The crystal structure of SalL revealed a channel which leaves the 2-position of ClDA solvent exposed. This raises the possibility of accommodation of 2-modified ClDA analogues as substrates of SalL. A suite of 2-modified ClDA analogues were synthesised to test this hypothesis. The current substrate scope of NovO is narrow so it was desired to investigate coumarins which have not yet been investigated with NovO. A range of coumarins were synthesised including a precursor to a compound which has shown activity as an inhibitor of Heat ShockProtein 90 (Hsp90) which has been implicated in a number of cancers. To explore the use of NovO as a method for late-stage functionalisation of drug molecules methylation of 7-hydroxywarfarin, a metabolite of clinically approved drug warfarin, has been carried out.

Proteolysis targeting chimeras (PROTACs) are heterobifunctional small molecules which induce targeted protein degradation by redirecting the ubiquitin-proteasome system. PROTACs simultaneously bind to both a protein of interest and an E3 ubiquitin ligase. The proximity of the target protein and the E3 ligase complex allows transfer of ubiquitin onto the target protein, after which the protein can be recognised and then degraded by the proteasome. The PROTAC mechanism-of-action offers a number of potential advantages over small molecule inhibition for the development of new medicines. Efficacy maybe achieved from low doses, extended duration of action is possible, arising from pharmacokinetic pharmacodynamic disconnects, and challenging targets may become tractable through the identification of suitable affinity binders. In this research, the development of PROTAC technology is explored. In order to expand the breadth of E3 ligases that are currently recruited using this approach, a promiscuous toolbox was established to prosecute new chemical matter for E3 ligases. In order to further elaborate the promiscuous toolbox already known for the degradation of kinases, investigation of a promiscuous bromodomain PROTAC was explored. After assessment of bromosporine derived PROTACs in multiple cell lines with two validated E3 ligases, it was deemed an unsuitable binder for the toolbox. The PROTACs were unable to induce potent nor promiscuous bromodomain degradation. As a result, a known BET bromodomain binder was selected for the promiscuous toolbox in addition to the known promiscuous kinase binder and a RIPK2 binder for new E3 ligase validation. With a promiscuous toolbox in hand, a new E3 ligase was evaluated. Indisulam, a small molecule "molecular glue", was found to bind to the E3 ligase DCAF15. Indisulam derived PROTACs were synthesised and evaluated with a range of linker lengths and multiple protein binders. These studies conclusively demonstrated that protein degradation was not achieved using these PROTACs. The indisulam derived binder was subsequently found not to be suitable for the PROTAC approach without further investigation to determine DCAF15 recruitment. Given the lack of degradation with the initial, empirically selected E3 ligases from the literature, a distinct E3 ligase agnostic approach to protein degradation was developed. A high throughput phenotypic screen was established using green fluorescent protein (GFP) as the protein of interest, where cellular fluorescence levels correlate with protein degradation. High throughput chemistry techniques were implemented and optimised to synthesise thousands of HaloCompounds in-situ by amide coupling. The compounds were tested directly in cells to find new chemical matter for the induction of protein degradation. This strategy allowed identification of several potential hits from a 3000-amine screen, with one high-confidence hit currently being further evaluated. For this effort the screen was optimised successfully and can potentially be employed for a target-agnostic high-throughput screening campaign of hundreds of thousands of compounds for new E3 ligases to employ in future protein degradation strategies.

Over half of the small molecules approved for use in the U.S.A. by the FDA in 2012 contained at least one nitrogen heterocycle. Of these, 11% contained a piperidine motif, emphasising the importance of saturated nitrogen-containing heterocycles in drug discovery. As such, rapid, step-efficient and reliable diversification strategies of this functional group are highly desirable. Late-stage C-H functionalisation of high value scaffolds is a powerful tool that has potentially wide application in the pharmaceutical industry. The underlying principle is to use the subtle difference in reactivity of C-H bonds to carry out selective and efficient functionalisation of a complex late-stage intermediate or final compound. This strategy could be harnessed to expedite the exploration of the medicinal chemistry properties of biologically active small molecules. There are a wide array of methods at the disposal of synthetic chemists to carry out selective C-H functionalisation, which are discussed within this thesis, but a number of these are not applicable to the late-stage functionalisation of modern drug molecules. This can be due to functional groups being incompatible with the reaction conditions, the inconvenience of installing an appropriate directing group, or the functionalisation not being sufficiently selective. This body of work provides an introduction to the field of late-stage C-H functionalisation. Chapter two describes the development and consolidation of a C-H functionalisation protocol for the α-C-H oxidation of cyclic amines to lactams using molecular iodine as the oxidant. This transition metal-free process, carried out under ambient reaction conditions, was then applied to the late-stage oxidation of a selection of high value small molecules with relevance to the pharmaceutical and agrochemical sectors. The impact of this work within a medicinal chemistry application is also discussed. With an understanding of this process in hand, diversification of this oxidative platform is explored in chapter three, which describes the discovery of a β-C-H trifluoroacylation protocol. Chapter four describes how these findings were refined and consolidated into the development of a robust methodology to carry out oxidative C-H sulfonylation of saturated nitrogen heterocycles at the β-position. The functionality installed into resulting enaminyl sulfone scaffolds was then exploited to provide access to a wide array of highly functionalised azacyclic scaffolds. Iodine-mediated oxidation is demonstrated to provide access to either α or β-functionalised products, which have been showcased to have application for medicinal and synthetic chemistry.

Pyrrole-Imidazole (Py-Im) hairpin polyamides are a programmable class of compounds that bind in the minor groove of double-stranded DNA (dsDNA). The hairpin conformationenables sufficient flexibility to form a side-by-side arrangement within the minor groove of target dsDNA sequences. An Im-Py pair discriminates G over C, A and T, whereas a Py-Pypair binds A and T over C and G. The possibility to target specific dsDNA sequences enables polyamides to modulate gene expression by the disruption of transcription factors such as the androgen receptor, over expressed in prostate cancer. A current limitation of the biological applications of Py-Im polyamides is their variable cell permeability profile. Im-richpolyamides tend to have reduced cellular and nuclear uptake in eukaryotic cells. Although modifications to the periphery of the polyamide scaffold have been explored at length, there has been limited studies on tuning the physicochemical properties of the G-recognising Imunit. This thesis describes the preparation and the evaluation of the binding mode of hairpin polyamides with Im units replaced by a series of 5-substituted 2-amino-4-carboxylic thiazole(Nt) building blocks. Chapter 1 introduces the endogenous and exogenous DNA recognition by protein and small molecules, the modulation of gene expression using Py-Im polyamides and the limitations of this approach. Chapter 2 describes the preparation of the building blocks and the optimisation of the solid phase synthesis of hairpin polyamides containing Nt units. These optimised conditions were used for the preparation of a suite of 8-ring polyamides incorporating Nt in different positions of the polyamide scaffold. Expanding the DNA binding repertoire of Pyrrole-Imidazole polyamides Chapter 3 describes the evaluation of the binding mode of Nt-containing polyamides using a combination of biophysical and biochemical methods. The 5-isopropyl substituted Nt (iPrNt) building block was identified as an alternative to Im for targeting G when incorporated at theN-terminus of the polyamide scaffold. Chapter 4 describes the structural characterisation of three selected polyamides binding to their dsDNA target sequence using NMR spectroscopy and molecular dynamics. This study reveals that the polyamide containing iPrNt at the N-terminus induces a greater compression of the major groove compared to the Im-containing analogue. Finally, Chapter 5 reflects on the work of this thesis and offers some perspective of the future development of Py-Im polyamides as gene expression modulators.

The Cu-catalysed alkyne-azide cycloaddition (CuAAC) or 'click' reaction is a powerful and robust bio-orthogonal reaction that exclusively produces 1,4-substituted triazoles. Despite its extensive utility in chemical biology, the ability to differentiate alkyne subtypes has received little attention as a tool for the construction of discrete bioconjugates. This work highlights the utility of aromatic ynamines as a new click reagent for sequential bioconjugation. Aromatic ynamines are superior click reagents with enhanced chemical reactivity relative to conventional alkynes. This unique and orthogonal reactivity profile circumvents the need for conventional protecting group strategies. This project will also highlight the biocompatibility of these reagents as a new tool for protecting-group free sequential CuAAC bioconjugation of oligonucleotides in the presence of more accessible competing alkyne substrate (Scheme 1). This strategy allows the formation of a new platform for specific labelling using fluorescent and PET probes as much as specific targeting and drug delivery. Importantly, higher reactivity of aromatic ynamines allows lower copper loading, thereby decreases toxicity and side reactions on biomolecules. [Graphic with the following title:] Scheme 1. Chemoslective, sequential CuAAC bioconjugation using enhanced reactivity of aromatic ynamines.

One of the most important post-transcriptional processes that occurs during the transcription of DNA to RNA is alternative splicing. Alternative splicing is a highly regulated process that occurs during gene expression and is the basis for the large protein diversity that is achieved in eukaryotes. It allows multiple protein isoforms to be formed from one pre-mRNA sequence. Although alternative splicing is necessary to provide protein diversity, aberrant splicing can also lead to diseases such as cancer. Recently, there has been an interest in the ability of small molecules to exogenously modulate alternative splicing as therapeutics for aberrant splicing diseases. The ellipticine analogue GQC-05 has been found to have splice switching capabilities towards the apoptotic oncogene Bcl-x, however the mode of action of GQC-05 is not understood. This thesis describes efforts made to elucidate the mode of action of GQC-05 by synthesising a library of ellipticine derivatives for testing in a range of biochemical assays to evaluate their ability to bind and stabilise Bcl-x. The background to alternative RNA splicing and the aberrant splicing process that can lead to diseases such as cancer is introduced in Chapter 1. The importance of the oncogene Bcl-x and the current limitations in targeting it therapeutically are also presented. Chapter 2 investigates current literature methods to prepare 9-methoxyellipticine. However it is found that these methods do not provide a modular framework in which to systematically alter the ellipticine structure. Chapter 3 describes the development of a one-pot, sequential, palladium cross-coupling methodology to construct the backbone of the natural product ellipticine. This methodology has been further extended to synthesise a range of ellipticine analogues with modular changes throughout the structure. Chapter 4 examines the ability of the ellipticine derivatives synthesised to stabilise a putative Gquadruplex structure within Bcl-x through the use of UV melting assays. The ligands have also been investigated for their ability to stabilise a range of competitor DNA structures with the help of FRET melting assays. The interaction of GQC-05 with the putative Bcl-x G-quadruplex has also initially been examined using NMR spectroscopy. And finally, the ellipticine derivatives have been tested in Bcl-x in vitro splicing assays to determine whether they are capable of exerting the same splicing modulation on Bcl-x as their parent compound GQC-05.

Regiospecific monoalkylation and fluoroalkylation of aromatic compounds is of key importance in medicinal chemistry, yet can be difficult to accomplish chemically. Enzymatic methods may offer a viable alternative to traditional C-C bond forming reactions, such as the Friedel-Crafts alkylation. This thesis describes a number of key advances in the use of two S-adenosyl-L-methionine (SAM) dependent methyltransferases (MTs) NovO (Streptomyces spheroides) and CouO (Streptomyces rishiriensis) as biocatalysts for the Friedel-Crafts alkylation. Firstly, NovO was expressed, purified and crystallised and the X-ray crystal structure of SelMet-NovO was solved to 1.9 Å resolution (PDB accession code: 5MGZ). Additionally, a homology model of CouO (85% sequence identity to NovO) was generated using the NovO structure. Both NovO and CouO are Class I MTs, comprising a Rossmann fold with additional α-helices at the C- and N-termini. Secondly, a catalytic mechanism for NovO was proposed based on the X-ray crystal structure, substrate docking studies, kinetic isotope effect (KIE) data and mutational analysis. This work identified a His-Arg motif to be central to the catalytic mechanism, with initial deprotonation by H120 and stabilisation of the resulting intermediate by R121. Additionally, the KIE data showed that methyl transfer was the rate limiting step. Finally, comparison with CouO identified position 117 in both proteins to be key for mediating the substrate scope of both enzymes. A third major contribution towards the development of NovO and CouO as biocatalysts was preliminary work on their directed evolution (DE) towards a wider substrate scope. Although a more efficient screening or selection strategy will be necessary to increase the substrate scope further, one mutant of NovO (N117M) was identified with higher activity toward 1, 6-dihydroxynaphthalene relative to the WT enzyme, serving as proof of concept for the 'evolvability' of these MTs. The development of a platform for biocatalytic fluoroalkylation using MTs was also explored, with extensive attempts towards the synthesis of a suitable fluoroalkylated cofactor, including the synthesis of novel fluorinated amino acids. A final key contribution towards the development of a biocatalytic Friedel-Crafts alkylation platform was the development of an in situ cofactor synthesis system using the halogenase Sa1L (Salinospora tropica). This was demonstrated on a preparative scale using crude cell lysates (methyl transfer) or purified enzymes (CD3, 13CH3, 13CD3 and Et transfer). This work addressed the issue of cofactor cost and instability associated with SAM dependent MTs.

In this thesis, two-dimensional infrared spectroscopy (2D-IR) spectroscopy is used to study changes in structure and dynamics in deoxyribonucleic acid (DNA) containing only Adenine-Thymine (AT) base pairs. The aims of the studies in this thesis are to demonstrate the ability of 2D-IR spectroscopy to extract unique dynamic information, not accessible via established experimental methods from nucleic acid systems, as it has for protein- and peptide-based systems. The underlying theory of both linear and nonlinear 2D-IR spectroscopies is described in the initial Chapter, following this the details of the types of information already obtained using these methods from DNA in the current literature is presented. Chapter (2) describes the experimental setups and methodologies used to acquire the data in this thesis. A description of FTIR spectroscopy as well as the two laser systems used to acquire 2D-IR data (ULTRA and LIFEtime) are presented. In Chapter (3), the thermal denaturation of an AT DNA sequence 15 base pairs in length is studied with 2D-IR in the spectral region where vibrational modes of the DNA bases are observed. Changes in the unique features observed in these spectra that are sensitive to DNA conformation are described along with the suitability of 2D-IR for accurately following the transition from double- to single stranded DNA compared to established methods. Changes in solvation dynamics experienced by functional groups of the DNA bases are also discussed. Chapter (4) demonstrates the ability of 2D-IR to explore multiple spectral regions simultaneously using a two-colour 2D-IR method. This allowed examination of communication between the vibrational modes of the DNA bases and those of the DNA backbone in the same DNA sequence used in the previous Chapter. Dynamics extracted from the features linking these two spectral regions present evidence for an energy transfer pathway through DNA that could be responsible for allowing dissipation of excess energy absorbed by DNA, preventing photo-damage of sequences. Chapter (5) expands on the studies in the previous Chapter by employing the same experimental methodology to AT DNA sequences of lengths varying from 15 bases pairs to a single base mononucleotide. The changes in observed features that link the base and backbone regions for the different sequences is presented along with the effects varying the sequence length has on the timescales of the energy transfer mechanism described in the previous Chapter. In Chapter (6) the conclusions presented in the Chapters prior to this point are drawn together to highlight progress made in gaining new insights into structure and dynamics of DNA and also presents possible future directions for studying nucleic acid systems with 2D-IR.

Deoxyribonucleic acid (DNA) is a fundamental component in all living organisms found in Nature. Although both the double-helix structure of this biomolecule and the fact that it stores all the genetic information required by the organism to survive are well understood there are still aspects related to this molecule which are not as clear. Particularly the interactions underpinning the formation of complexes between other molecules, such as proteins, and DNA remain unclear. One of these is the process underlying the formation of small molecule-DNA complexes. Such complexes are known to form via interactions between these small molecules and the DNA minor groove, it has proven to be complicated to develop a set of rational rules for the synthesis of binders to target particular DNA sequences. Such rules are complicated by the complex molecular environment found within the DNA minor groove as well as the role of the spine of hydration found within this groove. Another aspect of the behaviour of DNA which has attracted a lot of attention is related to the mechanism underlying the melting of short DNA sequence. It is important to gain a better understanding of these mechanisms as this process is thought to be important in DNA transcription, replication and repair as well as being important for the application and development of DNA scaffolds. Additionally, it is thought that understanding the impact of binding on this transition could be used to gain further insights into the interactions underpinning these complexes. Here, FT-IR, ultrafast IR pump-probe and two-dimensional infrared (2D-IR) spectroscopy have been applied to investigate the interactions underpinning the formation of small molecule-DNA. Utilising a specifically designed minor groove binding ligand, incorporating an azide moiety as a non-natural IR probe, the questions outlined above were addressed. The spectroscopy of the of azides and the DNA bases were initially studied separately, in order to gain the understanding of the spectroscopy of these vibrational modes necessary to maximise the information extracted from the DNA complexes. From the initial investigation of the asymmetric azide stretch of benzyl azide, a model compound, it was found that this mode could be used to determine the electrostatic potential and hydrogen bonding strength of the local environment. In the case of the DNA base modes, it was found that both changes in the structure of the duplex, due to alterations in the sequence and the binding of an archetypical minor groove binder, Hoechst 33258, could be reliably extracted. For the binding of Hoechst 33258, these modes reveal details about the interactions underpinning the formation of complexes with both its target and a sub-optimal DNA sequences. The insights gained were then brought together to study the complexes formed between DNA and the specially-designed ligand. This allowed both the interactions underlying these complexes and the nature of the water in the minor groove to be explored. Finally, these new spectroscopic methods were then utilised to begin to reveal details of the melting transition of these DNA duplexes and the impact on melting of ligand binding.

Tuberculosis (TB) remains one of the world's most lethal infectious diseases. The emergence and increasing prevalence of drug-resistant strains of Mycobacterium tuberculosis (M.tb) highlights the urgent need for new antitubercular medicines. Decaprenylphosphoryl-β-D-ribose 2'-epimerase 1 (DprE1) and N-acetylglucosamine-1-phosphate uridyltransferase (GlmU) are two recently characterised carbohydrate-processing enzymes essential for M.tb cell wall biosynthesis, both of which have promising potential as targets for new TB therapies. This thesis describes efforts to identify inhibitors of DprE1 and GlmU as possible new drug candidates for the treatment of TB, or as tool compounds to help develop a wider understanding of these novel targets. Chapters 1-3 introduce TB, physicochemical properties and their application in drug discovery, and DprE1 respectively. Chapters 4-6 detail hit-to-lead investigations around two compounds identified in a DprE1 high throughput screening (HTS) campaign. Hit validation was performed first, then each hit was expanded into a lead series, delineating structure-activity-relationships and identifying exemplars that displayed both high potency at DprE1 and against M.tb in vitro (with demonstrable engagement at DprE1). An overall aspiration was to identify DprE1 inhibitors that had high potential to succeed as new medicines for TB; as such, physicochemical properties were at the forefront of the molecular design strategy in these Chapters. Chapter 7 details the design, synthesis and biological evaluation of DprE1 inhibitors based on the natural substrate of the enzyme. This work focussed on investigating whether such ribose-based structures could provide a hydrophilic start-point from which to develop potent DprE1 inhibitors. These substrate analogues were also used as tool compounds to probe interactions between DprE1 and small molecule inhibitors, and to provide insight into the nature of the enzyme-catalysed reaction. Chapter 8 of this thesis introduces the second M.tb target, GlmU. Chapter 9 outlines the design and synthesis of a small set of GlmU acetyltransferase inhibitors based on the substrate of the enzyme. In the absence of any quality literature M.tb GlmU inhibitors, this work provided chemical matter to help establish the recently developed GSK M.tb GlmU acetyltransferase enzyme assay. Finally, Chapter 10 reflects on the work presented in this thesis, and offers some perspectives for the future of tuberculosis drug discovery.

In 2010 there were an estimated 219 million cases of malaria worldwide, accounting for approximately 0.66 million deaths. With growing resistance to current antimalarial agents a challenge is placed on the scientific community to provide efficacious and cost effective methods for the diagnosis and treatment of malaria. As protein phosphorylation regulates most aspects of cell life, understanding the function of protein kinases in malaria parasites has the potential to uncover novel drug targets. To this end, this work focused on the study of three essential Plasmodium falciparum kinases using two different chemical genetic approaches: γ-modified ATP analogues for the investigation of PfCK2; and covalent complementarity for the study of PfCLK1 and PfCLK3. The work presented here highlights the instability of the P-N bond of ATP phosphoramidates during the acidic conditions required for analysis by MALDI-TOF MS and seeks to overcome this through the development of alternative linkers to the γ-phosphate of ATP. Apparent IC[subscript 50] values recorded provide evidence of PfCK2α tolerating modification of the γ-phosphoryl group of ATP; however, no evidence was found to support GTP serving as an alternative co-substrate of PfCK2α. The gatekeeper mutant kinases PfCLK1F630C and PfCLK3F444C were successfully produced. It was found that this mutation rendered PfCLK3 inactive and was detrimental to the activity of PfCLK1. Under the conditions used PfCLK1F630C was not inhibited by a panel of 23 electrophilic inhibitors.

RNA splicing is a key process in gene expression and regulation in Eukaryotes and involves the processing of pre-mRNA sequences into mature mRNA. Pre-mRNA consists of exons (protein coding regions) and introns (non-protein coding regions). The introns of the pre-mRNA are excised and the exons ligate to form mature mRNA ready for export from the nucleus. Within the pre-mRNA there are numerous splice sites, some of which are conserved whilst others are alternative splice sites. RNA splicing can follow two different pathways: constitutive or alternative and in humans around 90% of pre mRNA is alternatively spliced which accounts for the formation of multiple isoforms of a single gene. The regulation of splicing involves cis-acting factors which are enhancer/silencer sequences within the pre-mRNA and trans-acting factors comprising of cellular factors including RNA and proteins, combined together they enhance or silence splicing. A major challenge in the field is to determine the interplay between the various factors associated with the promotion or silencing of specific splice sites. Two putative models for the utilization of splice sites have been proposed; firstly, a looping mechanism whereby the enhancers randomly collide with each other by 3D diffusion forming a loop. Secondly, enhancer mediated splicing occurs by a cooperative protein binding process. However, due to the limitations of current biochemical tools, it is not possible to determine the exact mode of action. In this thesis, a new chemical biological approach has been developed which addresses this question, involving the construction of tripartite RNA constructs separated by a non-RNA tether. Using these model systems, compelling evidence is provided which demonstrates splice site selection does not proceed via a looping mechanism which is the widely accepted model in the field.

Spinal muscular atrophy is the leading genetic cause of infant death, and much research has gone into the development of potential therapies for the disease. It is caused by a loss of the SMN1 gene. However, patients have the SMN2 gene, which contains a few silent mutations causing the skipping of exon 7 during splicing. One of the most promising therapeutic strategies involves the use of antisense oligonucleotides to rescue the splicing of SMN2 exon 7, allowing for production of full length SMN protein. One successful antisense oligonucleotide strategy involves the use of a bifunctional targeted oligonucleotide enhancer of splicing (TOES) (Skordis et al., 2003). This oligonucleotide, named GGA, consists of an annealing region that targets it to SMN2 exon 7 and a non-annealed enhancer tail domain, designed to recruit activator proteins and stimulate inclusion of exon 7. However, the precise mechanism by which GGA induces exon 7 inclusion was not fully understood at the time of design. This study has focused on investigation of the mechanism of action of GGA, in order to improve the therapeutic potential for GGA and future TOES. GGA was found to bind directly to SRSF1, an activator protein, via its enhancer tail domain. The tail domain forms a G-quadruplex structure in vitro (Smith et al., manuscript submitted). The presence of this structure in nuclear extracts was confirmed, and the enhancer domain was found to bind the G-quadruplex associated proteins CNBP and nucleolin. Stabilization of this structure using ligands reduced the efficacy of GGA, indicating that GGA does not form a G-quadruplex when it is actively stimulating SMN2 exon 7 inclusion. Single molecule methods revealed that the annealing domain of GGA, which anneals over an exonic splicing silencer shown to bind hnRNP A1 and/or Sam68 (Kashima et al., 2007; Pedrotti et al., 2010), reduces the number of Sam68 proteins bound per SMN2 RNA. These findings are consistent with the fact that the annealing region of GGA promotes U2AF65 binding and the enhancer tail domain promotes U2 snRNP binding to SMN2 transcripts (Smith et al., manuscript submitted; Smith, 2012).

In the current thesis we analyse protein phosphorylation pathways in P. falciparum, the protozoa responsible for the most virulent form of malaria in order to both understand the role and scope of this protein modification in the parasite, and to explore its feasibility as a new drug target. With the aim to map phosphorylation pathways controlled by P. falciparum Casein Kinase 2 (PfCK2), we developed a new chemical-biological approach based on γ-modified ATP analogues bearing reporting groups on the transferred phosphate in order to selectively tag CK2 substrates. Despite being able to efficiently synthesise a small set of analogues, the data presented here shows that the P-N linkage bond between the nucleotide and the tag is stable during the assay conditions but not during the product analysis due to its acidic liability (e.g. with HPLC, MALDI); suggesting that a different type of linkage should be chosen in the future. Detailed characterisation studies of the parasite PfCK2 presented here showed a number of important features differing from human CK2. Docking analyses with a CK2 inhibitor showed that the PfCK2 ATP binding pocket is smaller than human CK2 due to the presence of Val116 and Leu45 which in the human kinase are replaced by more bulky isoleucine residues: Ile120 and Ile49. The difference between the human and parasite CK2 orthologues extends further to mechanisms of activation and regulation. Shown here is the autophosphorylation of PfCK2 that, unlike the human orthologue, occurs within subdomain I at Thr63. This autophosphorylation is essential for full catalytic activity. In addition we also showed that Thr63 phosphorylation regulates the interaction between the calalytic α-subunit and the regulatory β2-subunit. Here, we also presented evidence for tyrosine phosphorylated proteins in parasite infected red blood cells. PfCK2 can act as a dual specificity kinase phosphorylating P. falciparum Minichromosome Maintenance protein 2 (PfMCM2) on Tyr16 in vitro. It is therefore possible that PfCK2 may contribute to tyrosine phosphorylation within the parasite. Finally, we also reported a study regarding MCM2-Ser13 phosphorylation which successfully identified PfCK1 as the kinase responsible for this event.

A key problem in nanotechnology is the integration of individual components into larger networks capable of more complex processes. DNA based photonic wires are a promising solution as they have been shown to transmit light energy over 10 nm distances, but are limited by their problematic assembly and reliance on fluorophore labelled DNA. This thesis describes efforts to construct an improved photonic wire using functionalised DNA binding small molecules as proof of principle for a ‘mix and match’ approach to nanotechnology which delivers individual components to a specific site on DNA. Polyamides have been shown to bind to DNA with very high affinity and specificity which together with their modular nature makes them an ideal ‘delivery system’. To combine this with the versatility and efficiency of copper catalysed click chemistry, novel internally functionalised alkyne polyamides were synthesised using both solution and solid phase chemistry. A general route to produce these internally modified polyamides was developed and the synthesis of the standard polyamide building blocks was improved. Test click reactions on alkyne polyamide fragments showed up to 92% conversion, but the same reactions failed on the full length polyamides and previously reported modification methods were used to create a fluorophore labelled polyamide. A coumarin based fluorophore was selected to allow direct substitution into proven photonic wires, but when the DNA binding affinity of this polyamide was tested, it was found that only weak binding was observed with 1.5 equivalents of polyamide. Upon construction, the improved photonic wire transported energy over a distance of 6 nm with an overall efficiency of 9% which was attributed to the poor DNA affinity. This poor performance makes it difficult to assess the general potential for this ‘mix and match’ approach, but the non-applicability of click chemistry and improvements in the synthesis will inform future designs.