BARBARA W. LOW AND JOHN T. EDSALL 



to dovetail into the space between neighboring residues on 

 another helix. Whether the formation of a coiled coil is attrib- 

 uted to such systematic steric effects or to more general polar 

 and nonpolar side chain residue interactions, the coiled coil 

 proposed has the advantage of explaining the observed dif- 

 fraction effects. 



The density of an a-helix array is, of course, dependent upon 

 the interaxial distance chosen. An interaxial distance of 

 approximately 10.6 A would give a reasonable density of about 

 1 .30. Such packing would give rise to a 9.2 A equatorial spacing. 

 Although 9.8 A is usually quoted as the a-pattern spacing, it has 

 been shown in well-resolved photographs that this broad 

 maximum is split into two reflections — with one at 9.2 A as 

 required. 



In a-keratin the chains are held together by cystine disulfide 

 linkages. The q:-/5 transformation in the k-m-e-f group of fibrous 

 proteins is usually considered to take place without — S — S — ■ 

 link breakdown. If the interchain cystine linkages are disposed 

 at random the a.-^ transformation could not take place without 

 breaking the disulfide linkages (6). Indeed simple steric 

 hindrance effects of neighboring side chains would appear to 

 make the transformation extremely difficult, even if the disulfide 

 linkages are located in positions favorable to forming the 

 stretched-out {0) configuration. This difficulty can be avoided 

 by consigning the — S — S linkages to the nonoriented regions 

 of the peptide chain. Polarized infrared radiation studies (2) 

 of keratin show rather low dichroism. They suggest that keratin 

 must contain considerable regions of unknown structure as well 

 as regions where the a-helix structure predominates. 



The evidence in support of the a-helix and ^ pleated sheet 

 structures as dominant features of the a and ^ fibrous protein 

 structure is in general diverse and formidable. Existence of the 

 a-helix structure has been clearly demonstrated in some synthetic 

 polypeptides. 



We may summarize the conclusions in rather diflferent 

 terms. The synthetic polypeptides examined so far have been 



406 



