Chemical Synthesis of Insulin Variants by KAHA Ligation

Abstract

Insulin is a protein hormone with a pivotal role in the regulation of numerous metabolic processes. Malfunction of insulin secretion or reduced sensitivity of the insulin receptor to its native ligand are the main reasons for diabetes mellitus. Recombinant technologies allow the large-scale production of some insulin variants, providing a treatment for millions of patients who rely on the daily administration of the hormone. The marketed insulin products are life-saving for the people living with diabetes, however they are not able to restore completely the healthy glucose metabolism. Mutations in the protein sequence or chemical modifications could improve the therapeutical properties of human insulin. Insulin is a small protein (51 amino acid residues), constructed from two peptidic chains (A- and B-chain) that are held together by two interchain and one intrachain disulfide bonds in its folded state. This unique structure makes insulin a challenging target for classical synthetic approaches. Inspired by the biosynthesis of insulin, we developed a general method for the chemical synthesis of insulin variants. We utilized a short, traceless prosthetic C-peptide for facile folding of linear insulin intermediates; the folding precursors were assembled by α-ketoacid–hydroxylamine (KAHA) ligation. The first part of the dissertation focuses on the development of the synthetic platform for insulin. The KAHA ligation proved to be very efficient in assembling the highly hydrophobic linear insulin variants. The applied prosthetic C-peptide made it possible to form of the disulfide bonds in a controlled fashion. Using our synthetic methodology towards insulin we generated four different variants of the protein (Glargine – M2, human, mouse, guinea pig) and were pleased to find that the synthetic proteins possessed comparable biological activities to a recombinant reference compound. The KAHA ligation based on the (S)-5-oxaproline monomer proved to be a powerful tool for protein synthesis. In order to further expand the scope of the reaction, new hydroxylamine monomers were investigated. In the following chapter we successfully applied the natural aspartic acid-yielding (S)-4,4-difluoro-5-oxaproline hydroxylamine monomer on the synthesis of a different Glargine – M2 insulin variant. We were pleased to see that the new monomer could operate under standard KAHA ligation conditions (DMSO/H2O with 0.1 M oxalic acid) and yield the ligated peptide in a comparable yield to the (S)-5-oxaproline monomer. The last part of the thesis describes the synthesis of an insulin variant containing a hydroxylamine moiety on the side chain of the incorporated ornithine residue that readily underwent potassium acyltrifluoroborate (KAT) ligation. By taking advantage of the incorporated hydroxylamine handle, we could access rhodamine-labeled, PEGylated and dimerized insulins by KAT ligation. The synthetic platform developed for insulin proved to be useful for the synthesis of several biologically active insulin variants – regardless the changes in the amino acid sequence. Furthermore the method enabled the incorporation of non-canonical amino acid building blocks for late-stage functionalization of the folded protein.