Niels C. Kaarsholm

The chromophoric divalent metal ion chelators 4-(2-pyridylazo)resorcinol (PAR) and 2,2',2"-terpyridine (terpy) are used as kinetic and spectroscopic probes to investigate in solution the SCN- -induced conformational transformations of the insulin, proinsulin, and miniproinsulin hexamers (miniproinsulin is a proinsulin analogue wherein the C-chain is replaced by a dipeptide cross-link between Gly-A1 and Ala-B30). Herein we designate the 2Zn and 4Zn crystal forms of the hexamer as the T6 and T3R3 conformations, respectively. For all three proteins, addition of SCN- reduces the rate of sequestering and removal of zinc ion by chelator. The effect of SCN- on the rate of this process saturates at the same concentration (30 mM) known to induce the T6 to T3R3 transformation in the insulin crystal. Under both T6 and T3R3 conditions, the critical stoichiometry for high-affinity interaction between Zn2+ and each of the three proteins is shown to be 2 mol of Zn2+/mol of protein hexamer. Consequently, we confirm the finding that off-axial coordination of Zn2+ via His-B10 and His-B5 residues is of minor importance for the SCN- -induced conformation change in solution [Renscheidt, H., Strassburger, W., Glatter, U., Wollmer, A., Dodson, G. G., & Mercola, D. A. (1984) Eur. J. Biochem. 142, 7-14]. Under T6 conditions, the kinetics of the reactions between insulin, proinsulin, and miniproinsulin and a variable excess of terpy are similar and biphasic.(ABSTRACT TRUNCATED AT 250 WORDS)

Insulin circulates in the bloodstream and binds to its specific cell-surface receptor as a 5808 Da monomeric species. However, studies of the monomer structure and dynamics in solution are severely limited by insulin self-association into dimers and higher oligomers. In the present work we use site-directed mutagenesis of the dimer- and hexamer-forming surfaces to yield the first insulin species amenable for structure determination at neutral pH by nuclear magnetic resonance (NMR) spectroscopy. The preferred insulin mutant, i.e., (B1, B10, B16, B27) Glu, des-B30 insulin retains 47% biological potency and remains monomeric at millimolar concentrations in aqueous solution at pH 6.5-7.5 as judged by NMR and near-UV circular dichroism (CD) spectroscopy. From a series of 2D 1H-NMR spectra collected at pH 6.5 and 34 degrees C, the majority of the resonances are assigned to specific residues in the sequence, and nuclear Overhauser enhancement (NOE) cross-peaks are identified. NOE-derived distance restraints in conjunction with torsion restraints based on measured coupling constants, 3JHNH alpha, are used for structure calculations using the hybrid method of distance geometry and simulated annealing. The calculated structures show that the major part of the insulin mutant is structurally well defined with an average root mean square (rms) deviation between the 25 calculated structures and the mean coordinates of 0.66 A for backbone atoms (A2-A19 and B4-B26) and 1.31 A for all backbone atoms. The A-chain consists of two antiparallel helices, A2-A7 and A12-A19, connected by a loop. The B-chain contains a loop region (B1-B8), an alpha-helix (B9-B19), and a type I turn (B20-B23) and terminates as an extended strand (B24-B29). The B1-B4 and B27-B29 regions are disordered in solution. The structure is generally similar to crystal structures and resembles a crystalline T-state more than an R-state in the sense that the B-chain helix is confined to residues B9-B19.

Insulin has a narrow therapeutic index, reflected in a small margin between a dose that achieves good glycemic control and one that causes hypoglycemia. Once injected, the clearance of exogenous insulin is invariant regardless of blood glucose, aggravating the potential to cause hypoglycemia. We sought to create a "smart" insulin, one that can alter insulin clearance and hence insulin action in response to blood glucose, mitigating risk for hypoglycemia. The approach added saccharide units to insulin to create insulin analogs with affinity for both the insulin receptor (IR) and mannose receptor C-type 1 (MR), which functions to clear endogenous mannosylated proteins, a principle used to endow insulin analogs with glucose responsivity. Iteration of these efforts culminated in the discovery of MK-2640, and its in vitro and in vivo preclinical properties are detailed in this report. In glucose clamp experiments conducted in healthy dogs, as plasma glucose was lowered stepwise from 280 mg/dL to 80 mg/dL, progressively more MK-2640 was cleared via MR, reducing by ∼30% its availability for binding to the IR. In dose escalations studies in diabetic minipigs, a higher therapeutic index for MK-2640 (threefold) was observed versus regular insulin (1.3-fold).

The insulin hexamer is an allosteric protein capable of undergoing transitions between three conformational states: T6, T3R3, and R6. These transitions are mediated by the binding of phenolic compounds to the R-state subunits, which provide positive homotropic effects, and by the coordination of anions to the bound metal ions, which act as heterotropic effectors. Since the insulin monomer is far more susceptible than the hexamer to thermal, mechanical, and chemical degradation, insulin-dependent diabetic patients rely on pharmaceutical preparations of the Zn-insulin hexamer, which act as stable forms of the biologically active monomeric insulin. In this study, the chromophoric chelator 2,2',2"-terpyridine (terpy) has been used as a kinetic probe of insulin hexamer stability to measure the effect of homotropic and heterotropic effectors on the dissociation kinetics of the Zn2+- and Co2+-insulin hexamer complexes. We show that the reaction between terpy and the R-state-bound metal ion is limited by the T3R3 <==> T6 or R6 <==> T3R3 conformational transition steps and the dissociation of one anionic ligand, or one anionic ligand and three phenolic ligand molecules, respectively, for T3R3 and R6. Consequently, because the activation energies of these steps are dominated by the ground-state stabilization energy of the R-state species, the kinetic stabilization of the insulin hexamer toward terpy-induced dissociation is linked to the thermodynamic stabilization of the hexamer. The mass action effect of anion binding and, foremost, of phenolic ligand binding provides the major mechanism of stabilization, resulting in the tightening of the tertiary and quaternary hexamer structures. Using this kinetic method, we show that the R6 conformation of Zn-insulin in the presence of Cl- ion and resorcinol is > 1.5 million-fold more stable than the T3 units of T6 and T3R3 and > 70,000-fold more stable than the R3 unit of T3R3. Furthermore, the stabilization effect is correlated with the affinity of the ligands: the tighter the binding, the slower the reaction between terpy and R-state-bound metal ion. These concepts provide a new basis for the pharmaceutical improvement of the physicochemical stability of formulations both for native insulin and for fast-acting monomeric insulin analogues through ligand-mediated allosteric interactions.

Abstract The aggregation properties of zinc‐free insulin have been studied using static and dynamic light scattering. The aggregation has been investigated as a function of three parameters, the concentration of sodium chloride (in the range 10–100 mM), the pH value (in the range pH 7.5–10.5), and the insulin concentration (1.8–13.4 mg/mL). The measured homodyne autocorrelation function was used to determine the apparent mean hydrodynamic diameter as well as the apparent weight‐averaged molar mass of the insulin species in solution. A method of data analysis was employed, which allows the separation of light scattering contributions from the insulin oligomers and from irrelevant macromolecules and possible impurities present in the sample solutions. Also, a simple phenomenological equilibrium model describing the association of oligomers of insulin is presented. One aspect of this model is that it makes it possible to determine weight average molar masses corrected for virial effects on the Rayleigh ratio. This was necessary because virial effects cannot be isolated and corrected for by dilution since this would change the equilibrium distribution of oligomers. The basis of the model is a positive contribution to Gibbs free energy from charge repulsion depending on the protein charge and the number of monomers in the oligomers, and an assumed constant negative contribution to Gibbs free energy arising from either an entropic gain or hydrogen bonding upon association. The equilibrium model gives a good description of both the apparent weight average molar masses and the apparent hydrodynamic diameters, when the effect of the insulin concentration is taken into account by including virial effects arising from charge–charge repulsion (Donnan effect). The result shows that the association of insulin as a function of pH and ionic strength can be described by an effective charge equal to the charge derived from proton titration reduced by the number of sodium ions binding to insulin. At the lowest pH and highest salt concentration (pH 7.5, 100 m M NaCl, 12 mg/mL insulin), the weight average molar mass is close to that of the hexamer, and at the highest pH and lowest salt concentration (pH 10.5, 10 m M NaCl, 1.9 mg/mL), the weight average molar mass is close to that of the monomer. In all cases, however, a distribution of oligomers is present with a relative Gaussian width of about 30%. Neglecting the positive term in Gibbs free energy, an upper bound to the association constant for insulin can be calculated: The negative term in Gibbs free energy corresponds to an association constant of (0.8 ± 0.3)·10 5 M −1 , which is in agreement with published values for the monomer‐to‐monomer association. The satisfactory agreement between theory and experiments for the weight average molar mass suggests that it should be possible to predict the aggregational properties of mutant forms of insulin. © 1993 John Wiley &amp; Sons, Inc.