Predictive modelling of peptide-based therapeutics: a way to accelerate biopharmaceutical design and formulation development

Peptide-based therapeutics are a unique class of biopharmaceuticals with distinct pharmacological properties and superior efficacy. Computer-assisted predictive modelling has revolutionised the design and development of peptide therapeutics. In this thesis, we have designed and optimised a wide array of natural and synthetic peptide analogues identifying design rules for new peptide therapeutics discovery that target protein receptors involved in metabolic disorders and cancers. Peptides are sustainable, biocompatible and tailorable next-generation alternatives to conventional therapeutics for G protein-coupled receptor (GPCR)- linked metabolic disorders including type-2 diabetes mellitus (T2DM), as agonists of class B1 GPCRs, glucagon receptor (GCGR) and glucagon-like peptide-1 receptor (GLP-1R). However, single agonist peptides that activate GLP-1R to stimulate insulin secretion may also suppress obesity-linked glucagon hormone release. Hence, peptide therapeutics co-targeting GCGR and GLP-1R may help remediate and tackle both diabetes and obesity to supersede current mono-agonist therapy. Here we design and model optimised peptide sequences starting from a phage-displayed peptide library screening. We first screen these peptides using predicted molecular binding profiles for dual agonism of GCGR and GLP-1R. Then we derive new peptide design rules from extensive molecular dynamics (MD) computer simulations based on peptide– GPCR binding profiles and report a unimolecular GCG/GLP-1 receptor co-agonist that could be experimentally developed to provide superior bodyweight management in obesity coupled with glycaemic control in T2DM. Our newly designed co-agonist peptide exhibits improved coupled affinity for GCGR and GLP-1R relative to both endogenous peptides and currently marketed drugs. In addition to GCG and GLP?1, experimental drugs are also developed based on the gut hormones such as glucose-dependent insulinotropic polypeptide (GIP), that control appetite and blood glucose. GIP, GLP-1 and GCG have some overlapping functionality, and their combination may lead to synergistic effects on diabetes and metabolic disease. Thus we further optimised the lead co-agonist peptide to equip it with triple receptor agonist activity, and trypsin and dipeptidyl peptidase-4 (DPP-4) enzymatic degradation resistance. With MD simulations, we explore peptide affinity and specificity towards three class B1 receptors including their potential for GPCR activation. Our latest designed tri-agonist peptide simultaneously targets GCG, GLP-1 and GIP receptors. Additionally, our models explore how the extracellular tri-agonist ligand binding facilitates the dynamics of downstream signalling by heterotrimeric G protein-coupling. Recently, there is significant interest in bioactive peptides as drug candidates from enzymatic hydrolysis of food proteins. We employed computational molecular modelling methods to screen and predict the binding and specificity of short biopeptides derived from milk proteins with improved multi-targeting affinity for the management of both T2DM and hypertension. Our rationally designed milk-derived tetra- and penta-peptides efficiently inhibit the DPP-4 enzyme, another therapeutic target for the management of T2DM by preventing the degradation of incretins (GLP-1/GIP). In parallel with DPP-4 inhibition, our modelled short peptides showed inhibitory potency for angiotensin?converting enzyme (ACE) for the management of hypertension. Finally, the cyclic peptide, somatostatin (SST) targets and stimulates class A GPCRs (somatostatin receptor subtypes, SSTR1-5). SST has anti-cell proliferative activity in neuroendocrine cancers/tumours by activation of SSTR2. Here, we report a cyclic SST-prodrug designed with an N-methyl nitroimidazole promoiety having an affinity for SSTR2. Our models predict greater specificity of SST in its prodrug form to SSTR2 compared to endogenous SST, highlighting improved delivery of SST-prodrug. Overall, this thesis provides novel molecular-level insights into multi?receptor targeting peptide therapeutics and GPCR dynamics via extensive computational molecular modelling techniques. The designed peptide therapeutics may accelerate biopharmaceutical formulation development for the treatment of cancers, and metabolic and cardiovascular disorders, including experimental peptide-based drugs targeting other GPCRs.