Molecular aspects of proinsulin C-peptide interactions

Abstract

Insulin biosynthesis in the beta-cells of the pancreas begins with the generation of preproinsulin, which is converted to proinsulin. Proinsulin is cleaved into equimolar amounts of insulin and connecting peptide (C-peptide), which are released into the circulation. The 31-aminoacid-residue C-peptide has been held solely to contribute to proper folding, disulphide bridge formation and processing of insulin, but many reports now suggest that it may be a hormone itself. It has a beneficial effect on glycaemic control and may protect against diabetic complications.

Comparison of the primary structures of proinsulin from 37 species showed that insulin A- and B-chains are considerably more conserved than C-peptide. Residues of Cpeptide conserved among mammalians reflect functional importance. It had been proposed that C-peptide biological activity involves association with lipid bilayers independently of chiral interactions, possibly involving its non-polar mid-section. The structure and capacity of human C-peptide to insert into lipid bilayers were assessed using circular dichroism and NMR spectroscopy. Lack of stable C-peptide secondary structure in aqueous solution was retained in the presence of lipid vesicles or micelles. C-peptide and lipid vesicles did not co-migrate upon size exclusion chromatography, which suggested that C-peptide is not likely to act via stable interactions with lipid membranes.

Specific binding of C-peptide to primary cells had also been presented, but a receptor specific for C-peptide had not been purified. Binding of rhodaminelabelled human C-peptide (Rh-C-peptide) to intact or detergent-solubilized human skin fibroblasts was studied with fluorescence correlation spectroscopy. The zwitterionic detergent CHAPS released macromolecules with maintained C-peptide binding capacity from skin fibroblasts and may be useful in future purification protocols. A biologically active pentapeptide consisting of the C-terminal five amino acid residues of C-peptide had been found to displace cell membrane-associated Rh-C-peptide.

The relative importance of each pentapeptide residue was assessed by determination of the displacing capacity of C-peptide analogues, and it was concluded that Glu-27 is critically involved in binding to cellular targets. The importance of conserved residues for biological activity of C-peptide was evaluated by incubation of mouse fibroblasts with physiological concentrations of C-peptide analogues. The capacity to induce phosphorylation of extracellular-signal regulated kinase (ERK) 1/2 correlates with the presence of residues Glu-3, -11 and -27 as well as non-helix-breaking residues in the N-terminal third of C-peptide, which forms a helical structure in trifluoroethanol. A tripartite model of C-peptide activity is suggested, since the acidic N-terminal part, the non-polar mid-section, and the partially conserved C-terminal part are all implied in biological activities.

Reported insulinomimetic effects of C-peptide may be related to similarities with insulin. However, C-peptide binding to or interference with insulin binding to soluble, truncated insulin receptor (IR) or IGF-1 receptor was not detected with surface plasmon resonance (SPR) technology, and did not affect signalling pathways mediated by IR subtypes A or B in hamster beta-cells. Insulin oligomerization was detected and characterized with SPR and mass spectrometry. C-peptide augmented insulinlinsulin interactions in solution and reduced the amount of insulin hexamers in gas phase, which may explain clinical findings that C-peptide reinforces the effects of insulin on glucose metabolism. The molecular and clinical data are compatible with a role of C-peptide to promote insulin disaggregation, which appears relevant where insulin concentrations are high, i.e. at the secretion and injection sites.