Conventional treatments for prostate cancer have significant limitations making it difficult to control the disease. Cyclopropabenzindoles (CBI) are more biologically potent, stable and synthetically accessible analogues of cyclopropapyrroloindole (CPI) anti-tumour antibiotics, such as duocarmycin-SA and CC1065. A polymeric prodrug carrying a CBI drug attached to the polymeric backbone through a PSA cleavable linker peptide has two modes of selectivity: activation by PSA and the EPR effect. To synthesise a 5-amino-seco-CBI analogue, 2,4-dinitronaphthalen- 1-ol gave di-Boc-1-iodonaphthalene-2,4-diamine in five steps (triflation, SNAr displacement with iodide, reduction (loss of iodine), protection and restoration of the iodine. For the amino-seco-CBI, it was important to discriminate between N2 and N4. Acidic removal of the Boc-group(s) resulted in deiodination. NMR investigations showed an unexpected Wheland-like cationic intermediate. N3 of naphthalene-1,3-diamine was selectively trifluoroacetylated and N1 was masked with Boc. Electrophilic iodination gave an orthogonally protected 1-iodonaphthalene-2,4-diamine. Allylation at the trifluoroacetamide was followed by free radical cyclisation with TEMPO trap. Removal of the trifluoroacetyl group allowed coupling to 5-(2-(dimethylamino)ethoxy)-1H-indole-2-carboxylic acid. Reductive removal of 2,2,6,6-tetramethylpiperidine, substitution of the exposed hydroxy group with chloride and removal of the Boc-group gave the amino-seco-CBI drug, 5-amino-1-chloromethyl-3-(5-(2-dimethylaminoethoxy)indole-2-carbonyl)-2,3-dihydro-1H-benz[e]indole. A DNA-melting assay confirmed that it binds very strongly to dsDNA causing a 13 deg. C increase in melting temperature. The drug was a highly potent cytotoxin in vitro, with IC50 = 18 nM against LNCaP prostate cancer cells. The polymeric prodrug system involved the synthesis of the pentapeptide SSKLQ. The amide side chain of glutamine can be masked as the nitrile and this can be quantitatively hydrated to the γ-carboxamide of L-Gln with hydroperoxide. The pentapeptide was coupled to 4-methoxynaphthalen-1-amine and to poly(ethylene glycol) as a model polymeric prodrug system. Efficient release of the model drug from the polymeric prodrug by PSA will allow this polymeric prodrug system to be adopted for the synthesised amino-seco-CBI drug.

The full text will be available at the end of the embargo period: 3rd April 2025 / The author's name as given on this thesis is Francis MPRAH BARNIEH. His publications use the name format Francis M. Barnieh.

<p>Therapeutic peptides have become a clinically and commercially important drug class providing novel treatment options in variety of disease areas. Today, more than 80 peptide drugs are marketed worldwide and hundreds more are in development. However, the development of peptide drugs can be hindered by their tendency to self-associate to form fibrils, an impurity that can affect potency and increase the potential for adverse immune responses in patients. Fibrillation of therapeutic peptides can present significant quality concerns and poses challenges for manufacturing and storage. From a pharmaceutical development perspective, early detection of instabilities can inform the development of mitigation strategies to minimize the risk of product failure and avoid costly delays in clinical development. A fundamental understanding of the mechanisms of fibrillation is critical for the rational design of fibrillation-resistant peptide drugs and formulations.</p> <p>The objective of this dissertation was to develop structurally modified fibrillation-resistant peptides based on a mechanistic understanding of the fibrillation process. The therapeutic peptides studied were human calcitonin (hCT), a glucagon/GLP-1 analog, and human insulin B-chain (INSB). Pulsed hydrogen-deuterium exchange mass spectrometry (HDX-MS) and other biophysical methods were used to provide mechanistic understanding of the intermolecular interactions and structural transitions during peptide fibrillation. Coupled with proteolytic digestion, pulsed HDX-MS of fibrillating peptides enabled identification of the residues involved in the early interactions leading to fibrillation based on their differential deuterium exchange rates. The high-resolution residue level information was used to make site-specific modifications to hCT, with phosphorylation in the central region resulting in complete inhibition of fibrillation for the phospho-Thr-13 hCT analog under the stress conditions employed. Reversible ‘prodrug’ modifications such as phosphorylation can aid the rational design of fibrillation-resistant therapeutic peptides. Furthermore, the effects of structural modifications on peptide fibrillation were evaluated by reducing the Cys1-Cys7 disulfide bond in hCT, and by C-terminal amidation or substitution with a helix-stabilizing residue (α-aminoisobutyric acid, Aib) in the glucagon/GLP-1 analog peptide. Finally, studies of insulin B-chain probed fibrillation mechanisms of this therapeutically important peptide, contributing to our understanding of the mechanisms of insulin fibrillation with the broad goal of developing fibrillation-resistant, rapid-acting, monomeric insulin analogs. Overall, the results demonstrate that small structural changes can have significant effects on peptide fibrillation, that pulsed HDX-MS can be used to probe these effects, and that an understanding of these effects can inform the rational development of fibrillation-resistant peptide drugs. </p>