Tricin (4',5,7-trihydroxy-3',5 '-dimethoxyflavone) from grass species and apigenin (4 ',5,7-trihydroxyflavone) from leafy vegetables are chemically closely related flavones. Quercetin (3,5,7,3',4 '-pentahydroxyflavone) is a flavonol contained in onions, wine, French beans and apples. These flavonoids have been reported to inhibit the growth of a variety of human-derived malignant cells in vitro at microM concentrations. Tricin, apigenin and quercetin showed differential growth-modulating abilities when they were incubated with colon-derived cells. Apigenin and more growth-inhibitory than tricin and quercetin in SW480 and HCA-7 cells. Whilst apigenin at 40 microM induced moderate apoptosis in HCA-7 cells, it arrested the cell cycle of SW480 cells without induction of apoptosis. In contrast to tricin and quercetin, which inhibited activity of COX enzymes strongly in a cell free system and decreased cellular PGE-2 levels moderately in HCA-7 and HCEC cells, apigenin lacked COX enzyme-inhibitory activity in the cell-free system. However, it reduced cellular PGE-2 moderately in HCA-7 cells and dramatically in HCEC cells, but it did not down-regulate COX-2 expression. The effect of apigenin on PGE-2 levels suggests that it may target the PGE-2 signalling pathway without effect on COX activity and expression. It is conceivable that apigenin affects cellular PGE-2 levels by induction or stimulation of the PGE-2 metabolising enzyme 15-hydroxyprostaglandin dehydrogenase. The results suggest that the pharmacological profile of apigenin renders it perhaps a more attractive chemopreventive agent than tricin or quercetin. The further exploration of the cancer chemopreventive properties of apigenin in rodent model in vivo is required.

The design, synthesis and biological evaluation of novel steroidal inhibitors of 17 β-Hydroxysteroid dehydrogenase type 1 which converts estrone into the highly potent estrogen, estradiol, is described. This isozyme has been implicated as a new drug target against hormone-dependent tumours which are stimulated by increased levels of estradiol. Initial targets were planned, around modifications of the steroidal estrone / estradiol core, utilising Mannich, Friedel-Crafts and amide coupling chemistry. When the biological activity of these compounds was evaluated in a cell-based 17β-HSD1 assay, significant inhibitory action was observed. The most potent compound was (13S)-3-hydroxy-13-methyl 2-(morpholinomethyl)-7,8,9,11,12,13,15,16-octahydro 6H-cyclopenta[a]-phenan-thren-17(14H)-one 128 with an IC50 of 723 nM, 128 was also shown to be selective for 17βSHD1 over 17β-HSD2 and does not display any cytotoxicity at 50 μM over a 96 hour period in estrogen receptor positive and estrogen receptor negative cells. A C16 extended linked and reverse amide series were synthesised, for which design was based upon STX1040 1 the most potent inhibitor in the literature. The most biologically active compound was N-(((13S,16R,17S)-2-ethyl 3,17-dihydroxy-13-methyl-7,8,9,11,12,13,14,15,16,17-decahydro 6H-cyclopenta[a]-phenan-thren-16-yl)methyl)-2-(pyridin-3-yl)acetamide 186 with a percentage inhibition of 81% at 10 μM. Similar biological activities were observed for the rest of this series, however these values are approximately 20% lower than those observed for STX1040 1. Quantities of H6-17β-HSD1 were successfully synthesised and purified for use in development of a novel MTS/PES recombinant protein assay, to assess the inhibitory potential of the novel steroidal compounds described and their modes of binding with the active site of 17β-HSD1. For these experiments it can be tentatively concluded that STX1040 1 acts a mixed site inhibitor for both the E2 and co-factor binding sites and the 2-substituted Mannich compound 127 acts as a competitive inhibitor for the E2 binding site and a non-competitive inhibitor for the co-factor binding site, however further experimentation is required to confirm these preliminary results.

The work described in this thesis details the design, synthesis and evaluation of molecules intended as mechanistic probes to the enediyne anticancer antibiotics. In particular it examines the role of radical and ionic accepting groups in analogue systems related to the DNA cleaving agent Neocarzinostatin chromophore (NCS). The application of novel methodologies towards the synthesis of a number of key precursors and model systems is discussed. Initial studies concentrated on the versatile synthesis of a model system bearing dienediyne and styryl motifs, from 2-cyclopentenone. The role of cyclic amine bases in a critical Stille type cross-coupling was investigated, leading to an efficient and robust synthetic route. Attempts to selectively cyclopropanate, or epoxidise, the styryl olefin under a range of conditions however, failed to give the desired phenylcyclopropyl or phenyloxiranyl analogues. A novel application of the Kulinkovich reaction, to form an alkynylcyclopropanol is discussed. Attempts to synthesise a phenylcyclopropanol analogue via an advanced tris-silyl pre-cursor are also detailed. The successfulu se of Salaun's cyclopropanone-hemiacetatl o complete the synthesis of a novel analogue system bearing a cyclopropanol motif is detailed and discussed, with some mention of related synthetic endeavours. The analogue's reactivity to a range of nucleophiles, radicals and reaction conditions is evaluated, though no conclusive evidence for cycloaromatised products was found. Finally, approaches to the synthesis of a fully functionalised analogue system incorporating dienediyne, cyclopropanol and selectively protected trans-diol moieties (for possible coupling to DNA binding agents) are detailed. Although this synthesis incorporates some methodology previously reported, significant improvementsnotably the synthesis of a key pyrano-3-one directly from furfuryl alcohol using Nbromosuccinimide and aqueous sodium bicarbonate- make it noteworthy of inclusion

The pyrrolo 2,1-c 1,benzodiazepines (PBDs) are a family of naturally occurring antitumour antibiotics which includes anthramycin, DC-81, tomaymycin and sibiromycin. They exert their biological activity through covalent binding to the exocyclic N2 group of guanine in the minor groove of DNA and block transcription in a sequence-specific manner. These PBD monomers span three DNA base pairs and have a preference for binding to purine-G-purine triplets. The PBDs have been used as a scaffold to attach other moieties, leading to novel sequence-selective DNA minor groove alkylating agents. In addition, as part of a rational approach to producing more efficient and selective DNA interstrand crosslinking agents, two PBD monomers have been linked together to form PBD dimers. The research in this thesis is a study of the molecular and cellular pharmacology of several series of novel PBD-containing agents including novel PBD dimers with different linker lengths, PBD-nitrogen mustard conjugates, PBD-polyamide conjugates and C2-aryl PBD monomers. Cytotoxicity in human tumour cell lines, efficiency of DNA interstrand crosslinking in naked linear plasmid DNA, and DNA sequence specificity were assessed. DNA interstrand crosslink formation and repair in cells were also measured. Resulting from this work Q2lQT-exo-unsaturated PBD dimers have been characterised to be highly cytotoxic and efficient in producing interstrand crosslinks both in naked DNA and in cells that are not repaired up to 48 hours. Only two of the PBD-nitrogen mustard conjugates showed some interaction with DNA although several members of this group showed significant cytotoxicity. A PBD-tri-pyrrole conjugate was found to bind preferentially to the sequence 5'-AGATTATC. Novel C2-aryl PBD monomers were shown to bind selectively to 5'-purine-G-purine sequences and demonstrated significant cytotoxicity. In addition, a method utilizing fluorescently end-labelled oligonucleotides was developed and validated to screen libraries of PBD-containing molecules synthesised on beads by combinatorial chemistry. This method allowed the isolation and discrimination of beads containing compounds, which have a high affinity for specific DNA sequences.

The A83586C/GE3 class of cyclodepsipeptides are a family of antitumour antibiotics whose antitumour properties have been attributed to their ability to inhibit E2F transcription factors. The latter are critical regulators of cellular proliferation that are potentially important new therapeutic targets for the control of proliferative diseases such as cancer. In this thesis, the asymmetric total synthesis of several A83586C/GE3/Verucopeptin analogues is described. Some of these molecules have provided valuable insights into the actual mechanism of antitumour activity for this class. All of these molecules have been built through a chemoselective coupling strategy involving the fully elaborated pyran /V-hydroxybenzotriazole activated ester 47 and the relevant unprotected cyclohexadepsipeptide. The latter were each synthesised through a 2+2+2 -fragment condensation strategy and macrolactamisation was accomplished with HATU. The approach used is exemplified below by our synthesis of the A83586C/GE3 hybrid 257.

Advances in cancer therapy have led to the exploitation of the biological differences between cancer and normal cells. A popular approach is to attach a targeting moiety such as an antibody to a toxin or toxic enzyme in order to specifically locate and kill cancer cells only. However, many immunotoxins have fared poorly in their clinical trials, having shown to elicit immunogenic responses and toxicity in the patient. This problem is commonly attributed to the non-human sources of the toxins and enzymes being delivered. Targeting cytotoxic agents from human sources is therefore a prospective solution. Accordingly, this project aims to develop an immunoconjugate using mammalian or 'mammalian-like' enzymes for the treatment of epithelial cancers over-expressing the MUCl tumour antigen.

Histone deacetylase inhibitors have great potential as chemotherapeutic agents both alone and in combination with other agents, however their mechanism of action is poorly understood. Studies into the effects of HDI on numerous tumour cell lines demonstrated some variation in the responses to these agents and gene expression profiling revealed that a variety of cellular processes are altered after HDI treatment.

Inhibition of the ST AT3: ST AT3 protein-protein interaction is an attractive approach for cancer therapy as it can lead to suppression of tumour cell growth and induce apoptosis. The racemic ochromycinone (STA21) is one of the few known small-molecule STAT3:STAT3 inhibitors. Our synthetic efforts focused on synthesis of the natural product YM-181741, which possesses at least three points for chemical variation to prepare compound libraries as potential STAT3:STAT3 inhibitors. Synthesis of the angucycline molecule was achieved employing an approach based on a Gold (lII)-catalysed intramolecular [4+2] benzannulation reaction. A facile and highly efficient route for the preparation of racemic form of the natural product was developed that offers a high degree of flexibility for modification of the scaffold at different stages of its synthesis. The enantioselective synthesis of (S)- YM181741 was successfully carried out through the (R)-diyne building block via an enantioselective copper-catalysed 1,4-conjugate addition reaction on a system bearing a v-coordinating group, in order to install the chirality on the diyne moiety. The optimised reaction conditions afforded the Michael adduct in good yield and high enantiomeric excess (up to 96% ee). Further chemical elaboration of the (±)- YM-181741 natural product was investigated in order to explore the necessary chemical diversity to assess a preliminary model of interaction between the SH2 domain of ST AT3 and the angucycline scaffold. The biological evaluation of the focused library allowed the identification of angucycline derivatives possessing high binding affinity for the SH2 region and the ability to inhibit STAT3 transcriptional activity.

Trabectedin (ET-743, Yondelis) is a marine alkaloid isolated from the tunicate Ecteinascidia turbinata, cytotoxic against a variety of tumor cell lines in vitro and human tumor xenografts in vivo. It has been approved by EMEA for the 2nd line therapy of soft tissue sarcomas in 2007 and for the 2nd line therapy of ovarian cancer in 2009. Myxoid liposarcoma (ML) is a specific histological subtype within the family of adults soft tissue sarcomas. Specifically >90% of usual myxoid/round cell liposarcomas (MLS/RCLS) are characterized by the chromosomal translocation t(12;16) (q13; pI I), which produces the FUS-CHOP oncogene. Different chimera subtypes seem to respond differently to trabectedin in clinical setting. To elucidate the mechanisms behind the differential sensitivity to trabectedin, tumor myxoid liposarcomas type II and type III were xenografted in nude mice, treated with trabectedin (0.15 mg/kg was injected i.v.) and the binding of FUS-CHOP to some of its target promoters was monitored by Chromatin immuno precipitation to verify the drug ability to displace the binding. We found that trabectedin was more effective on type II than type In ML xenografts. The response to trabectedin in Type II xenografts was associated with partial regression and pathological response. Type III ML xenografts appeared less sensitive to trabectedin and tumors did not regress unless a more prolonged treatment is performed. Molecular analysis revealed that trabectedin was able to remove FUS-CHOP type II and III from its own gene targets 24 hours after treatment. 72 hours after treatment, FUS-CHOP Type III was attached to its targets whereas FUS-CHOP Type II remained unbound. The results suggest that the difference in sensitivity of type II and type III ML tumors to trabectedin is related to a difference in duration of the drug's ability to detach FUS-CHOP from DNA and that type III ML required more intensive and prolonged treatment to achieve a sufficient respond to trabectedin.

One of the most important mechanisms for multidrug resistance (MDR) phenotype is believed be the P-gp-mediated drug efflux. The consequent lowering of the intracellular concentrations of many commonly-used chemotherapeutic drugs, such as doxorubicin (Dox), has been addressed by P-gp inhibition via covalent attachment of MDR drugs to carriers.This project aimed to develop new drug binding peptide structures able to traverse cell membranes and to investigate their potential in the 'non'-covalent' deiivery of Dox in drug-resistant (KD30) cells.