PROTEIN STRUCTURE 



Study of the globular proteins, in which they concluded that the 

 oil left-handed helix is the most plausible predominant chain 

 configuration. They write, "Our earlier suggestion (4) that the 

 insulin molecule is made up of equal numbers of a^ and ao 

 helices remains hypothetical and has not been capable either of 

 substantiation or disproof in the present work." 



Calvin (21a) has discussed the stereochemistry of the 

 — Ci — Si — S2 — Co — linkage. The position of minimum 

 potential energy corresponds to a configuration in which 

 the dihedral angle between the two — S — C bonds is nearly 

 90°. In other words, the plane defined by the Ci, Si, and S2 

 atoms is nearly perpendicular to the plane defined by Si, So, 

 and C2. The resistance to free rotation around the S — S bond 

 is very great ; surprisingly great for what is ordinarily considered 

 a single bond. This high potential energy barrier must be of 

 great importance in determining the configuration of the insulin 

 molecule in the regions of inter and intra chain disulfide bridges. 

 When the cystine linkage is in the favored configuration it will, 

 by Calvin's estimate, be more stable, by 10 to 20 kcal. per mole, 

 than the cis or trans configuration in which all four atoms lie in 

 the same plane. In the Lindley and Rollett model all three 

 cystine Ci — Si — S2 — Co linkages have the favored dihedral 

 angle between the S — C bonds. However, deviations from the 

 favored configuration do occur. In the crystal structure of 

 iV,iV'-diglycyl-L-cystine (110a) a dihedral angle of 101 ° =1= 2 

 has been reported. 



Both the Low and the Lindley and Rollett models can be 

 stacked in close-packed array. Lindley and Rollett have given 

 a detailed discussion of the possible modes of intermolecular 

 packing in the rhombohedral crystals of zinc insulin. They 

 suggest that the 12,000 unit, which is the stable monomer in 

 solution, may be made up of 2 (A + B) units packed in rec- 

 tangular array and held together by specific interaction between 

 zinc ions and histidine and glutamic acid residues. This 

 particular packing model unit appears rather unlikely as zinc 

 is not necessary to hold the 12,000 unit intact. Other packing 

 models of 2 (A + B) units can be devised. 



423 



o 



