From careful analysis of structure–function relationships of both naturally occurring and chemically modified insulins, it is apparent that certain structural features of insulin have been conserved throughout evolution. These include the following: (i) the precise positions of the three disulfide bonds; (ii) the N- and C-terminal regions of the A chain; and (iii) the hydro phobic residues of the C-terminal region of the B chain. From biological studies, it has been shown that insulin binds to receptors from tissues of differing species in a way that supports the view that there has been little or no evolutionary drift in the hormone binding site specificity of the insulin receptor. The insulin receptor specifically recognizes on the insulin molecule an extensive region of ~11 residues around residues A-21 and B-23–B-27. Also insulin binding to its receptor is dependent upon its intact tertiary structure.
Since the amino acid sequence of the evolutionary ancient hagfish insulin stabilized ~300 million years ago and differs from that of human at only 19 of 52 residues (37%), it can be calculated that amino acid substitutions have occurred in insulin at a rate of ~1 × 10−9 site/year. This is a typical figure for a highly conserved protein. These findings also imply that the insulin membrane receptor itself has not undergone significant changes throughout the course of vertebrate evolution.
D. Crowfoot Hodgkin, T. Blundell, and associates have determined via X-ray crystallographic techniques the three-dimensional structure of porcine insulin initially at 2.8-Å resolution and then at 1.5 Å. A highly similar structure of insulin in solution has been achieved via nuclear magnetic resonance (NMR) spectroscopy.
Human insulin is comprised of 777 atoms [C254H377,N65O75S6] with a molecular weight of 5,808 Da. Insulin can exist as a monomer; the left image is a “ball and stick” three-dimensional presentation that includes all of the atoms of the insulin monomer, while the right panel is a ribbon presentation of insulin which highlights the polypeptide backbone. In addition insulin also can spontaneously organize itself as a dimer, tetramer, and hexamer of the monomer. The insulin dimers in the crystalline state are held together by hydrogen bonds between the pep tide groups of residues B-24 (phenylalanine) and B-26 (tyrosine), which form an antiparallel pleated sheet structure. The hexameric structure of crystalline insulin consists of three dimers ordered around a major three-fold axis containing two zinc atoms that are each coordinated at the imidazole groups of three B-10 histidine residues. Indeed, the propensity for insulin to crystallize into hexamers may be related to the regular arrays of insulin that can be seen via electron microscopic evaluation of the storage granules of the pancreas β-cell.
The biosynthesis pathway for insulin generates monomers; the biologically active form of insulin is the monomeric species. However, monomeric insulin does readily spontaneously polymerize to form hexameric insulin, which is a functional long-term storage form of insulin.
While for many years it was thought that insulin was not a member of any hormone family, either in a structural sense (e.g., oxytocin and vasopressin are structural analogs of one another) or in a protein- processing sense (proopiomelanocortin produces β-lipotropin, ACTH, endorphins, etc.), there is now clear evidence that there is a hormone family of homologous growth factors that includes proteins with regions of amino acid sequences identical to that of insulin. These include the following: (i) relaxin, a polypeptide hormone from the corpus luteum that is responsible for the dilation of the symphysis pubis prior to parturition; (ii) insulin-like growth factors (IGFs) I or II; and (iii) nerve growth factor (NGF).
