Cyclic peptide therapeutics fill the gap between small molecules (5000 Da) as a separate class of drugs, combining the advantages of both in terms of high selectivity, bioavailability, synthetic accessibility and low toxicity. However, their conformational flexibility and proteolytic instability results in fast metabolism. Therefore, modification of their chemical and structural properties at the proteolytically vulnerable sites can circumvent these deficiencies. Several strategies exist to stabilise peptides through local or global constraints, side chain functionalisation or amide bond surrogates. This thesis describes the development and application of two chemical transformations to peptides with the aim of greatly enhancing their metabolic stability either via side chain to side chain cyclisation or disulfide modification. Following the functionalisation, a comparative evaluation of the pharmacological properties of the modified peptides was conducted with respect to the parent peptides. The first cyclisation strategy aimed to apply the redox-neutral chemistry developed in the burgeoning field of borrowing hydrogen technology to peptides, as a new class of tolerable substrates. These would contain the requisite amine and alcohol functional groups enabling their direct coupling to form the lactam or N-alkylated cycle. Numerous investigations into this transformation with a plethora of substrates and catalysts did not lead to the desired cyclisation. For the second approach, a simple one-pot procedure for the 'stapling' of disulfide bonds in native peptides was developed. Tris(2-carboxyethyl)phosphine mediated reduction of the disulfide bond was followed by concomitant insertion of a 'doubly electrophilic' carbon bridge to form a physiologically stable dithioacetal bond. The reaction was performed under mild, biocompatible reaction conditions, allowing for the facile conversion of native peptides into more stable analogues. The protocol was applicable to a range of bioactive peptides with a multitude of reactive functional groups demonstrating the high versatility of this approach. The importance and utility of the modified analogues were verified by testing their binding affinity and biological activity against the corresponding parent peptides. The impact of the disulfide modification on these properties for each peptide were analysed on a case-by-case basis. The SCS modified analogues of Oxytocin and Vasopressin maintained binding affinities in the same order of magnitude as their respective parent compounds across all corresponding receptors. Nevertheless, this was not the case for SCS-Octreotide and SCS-Somatostatin. Those peptides which maintained comparable binding affinities were then tested for their functional activity at the receptors of biological interest. SCS-Oxytocin retained its agonist potency displaying sub-nanomolar activity at the Oxytocin receptor, whilst SCS-Vasopressin exhibited sub micromolar functional response at the Vasopressin 1A receptor. Then the stability of the modified peptides in human serum was evaluated. In some cases, the enhancement in serum stability was vastly increased. In particular our modified SCS-Oxytocin analogue clearly demonstrated a significant improvement in serum half-life, as well as heat and pH stability over its native parent form. This disulfide stapling method could have wide-reaching application for the development of more stable therapeutic analogues.
EPFL2016
