Chemical reactivity of the functional groups of insulin. Concentration-dependence studies

Abstract

A modification to the competitive labelling procedure of Duggleby and Kaplan [(1975) Biochemistry 14, 5168-5175] was used to study the reactivity of the N-termini, lysine, histidine and tyrosine groups of insulin over the concentration range 1 X 10(-3)-1 X 10(-7)M. Reactions were carried out with acetic anhydride and 1-fluoro-2,4-dinitrobenzene in 0.1 M-KCl at 37 degrees C using Pyrex glass, Tefzel and polystyrene reaction vessels. At high concentrations all groups had either normal or enhanced reactivity but at high dilution the reactivities of all functional groups became negligible. This behaviour is attributed to the adsorption of insulin to the reaction vessels. The histidine residues show a large decrease in reactivity in all reaction vessels in the concentration range 1 X 10(-3)-1 X 10(-5)M where there are no adsorption effects and where the reactivities of all other functional groups are independent of concentration. With polystyrene, where adsorption effects become significant only below 1 X 10(-6)M, the reactivity of the phenylalanine N-terminus also shows a decrease in reactivity between 1 X 10(-5) and 1 X 10(-6)M. In 1 M-KCl insulin does not absorb to Pyrex glass and under these conditions the histidine reactivity is concentration-dependent from 1 X 10(-3) to 5 X 10(-6)M and the B1 phenylalanine alpha-amino and the B29 lysine epsilon-amino reactivities from 5 X 10(-6) to 1 X 10(-7)M, whereas the reactivities of all other groups are constant. These alterations in reactivity on dilution are attributed to disruption of dimer-dimer interactions for histidine and to monomer-monomer interactions for the phenylalanine and lysine amino groups. It is concluded that the monomeric unit of insulin has essentially the same conformation in its free and associated states.

Structure-function relationships in the free insulin monomer

“…Kaplan, 1979;Chan et al, 1981) demonstrate that, in the associated forms of insulin, the glycine N-terminus has a 'normal' pKa and a 'normal' reactivity on the basis of a Bronsted plot. From the examination of concentration-dependence of reactivity at pH 7.5 (Kaplan et al, 1984) one would predict that chemical properties of this group are the same in the free and associated states of the monomeric unit. Isolation of Dnp-glycine after competitive labelling of insulin with Dnp-F at 0.5 /lM bears out this prediction since the relative reactivity determined is the same as in the associated forms of insulin (Table 1).…”

“…An important question is whether the conformation of this species in solution is similar to or different from that of the monomeric unit in the hexameric form as determined by X-ray crystallography. There is very little direct evidence in the literature on this point, but the data that do exist (Wood et al, 1975;Chothia et at., 1983;Kaplan et al, 1984) indicate that the structures are probably very similar. The evidence, however, is far from conclusive on this point, since evidence for substantial differences has been reported (Pocker & Biswas, 1980).…”

“…The reactivity (acr) at each pH value was calculated from the following equations (Kaplan et al, 1971;1984):…”

“…10-5 M (Jeffrey & Coates, 1966;Pekar & Frank, 1972;Goldman & Carpenter, 1974;Milthorpe et al, 197-7k. This means that studies on the free monomer must be carried out at concentrations below 1 aLM. An addedt complicatioa is the fact that insulin in dilute solution strongly adsorbs on glass surfaces (Hollenberg & Cuatrecasas& 1976;Kaplan et al, 1984). Recent improvements in the competitive labelling methodology (Kaplan et al, 1984;Hefford et al, 1985) have made it possible to determine the reactivity of functional groups of proteins in extremely dilute solutions, so that the study can be carried out at a concentration (0.5 /tM) where the free insulin monomer predominates.…”

“…An addedt complicatioa is the fact that insulin in dilute solution strongly adsorbs on glass surfaces (Hollenberg & Cuatrecasas& 1976;Kaplan et al, 1984). Recent improvements in the competitive labelling methodology (Kaplan et al, 1984;Hefford et al, 1985) have made it possible to determine the reactivity of functional groups of proteins in extremely dilute solutions, so that the study can be carried out at a concentration (0.5 /tM) where the free insulin monomer predominates. The problem of adsorption can be eliminated by carrying out the study in 1 m-salt, where insulin does not adsorb on glass surfaces (Kaplan et al, 1984).…”