Glucagon is a polypeptide of 29 residues, which can be crystallised. Optical rotatory dispersion in dilute neutral, acid and alkaline solution suggests that the polypeptide chain is largely random, but contains about one turn of a-helix, which is eliminated in 6 M guanidine hydrochloride.Using thermal difference spectra, nuclear magnetic resonance, and the temperature-dependence of the optical rotatory dispersion as criteria, it appears that glucagon possesses no defined tertiary structure in solution, but may be regarded as bcing in a state ofa-helix s random coil equilibrium, near the high-temperature end of the transition range. That there is no steric objection to the formation of a highly a-helical conformation is demonstrated by optical rotatory dispersion measurements in 2-chloroethanoi solution, in which the protein becomes largely helical. A rough calculation indicates that in aqueous solution a t pH 2 the transition mid-point is at about -40". From changes in the proton magnetic resonance spectrum when glucagon is transferred from an aqueous environment into 6 M guanidine hydrochloride, it is deduced that the short helical segment in the chain in the aqueous state contains most of the long aliphatic side chains, and since the sequence is known it has been possible to identify such a segment uniquely near the C-terminal end. On standing in acid solution the viscosity increases and a birefringent gel is formed. Sedimentation studies indicate the formation of large aggregates. On further standing, or warming, a precipitate appears which has the appearance of long fibrils in the electron microscope. Infrared spectra of the gel, of solid films and of the precipitated material show that in all these states the glucagon is in the form of antiparallel P-chains.Glucagon is a peptide hormone of molecular weight 3,500, with a known sequence [I]. It has been reported [21 that glucagon crystals are cubic, and it was suggested that the molecule in the crystal is cylindrical, and t,herefore in the a-helical form. The only molecules of comparable length which are known to exist in the fully helical state are homopolypeptides, and a considerable body of evidence has built up [3,4] which indicates that the helical segments in globular proteins depend for their conformational stability on hydrophobic contacts in the interior of the protein. Moreover no proteins in the size range of glucagon have so far been found to have rigid tertiary structures. It is of interest t o investigate the secondary structure of glucagon under various conditions, in order to determine the nature of the secondary structure, and whether a tertiary structure can be said to exist. A brief account of part of the work on the aggregated states has appeared [5]. EXPERIMENTAL PROCEDURE MaterialsThe glucagon samples used in this study were given by Dr. J. Schlichtkrull (Novo Research Institute) and Mr. T. Couling (Lilly Research Laboratories Ltd.). This crystalline material was homogeneous by paper electrophoresis a t several pH values. I n starch gel ...