256 Walter Gordy 



link would of course tend to attract other agents such as O2 or H2O which 

 might later sever the bond or an electron which would restore the normal 

 S— S hnk. 



Although we are not yet certain whether the cystine or cysteine-hke resonance 

 arises from radicals of the type R — (S • • • S)+ — R or RS-, we are inclined to 

 favor the latter. It would seem that the neutral radical would enjoy the longer 

 life and hence be the more probable one to be detected. Furthennore, in the 

 RSH compounds the formation of the RS- radical would require the simpler 

 process. With the present information we are inclined to beheve that 

 R — (S • • • S)+ — R is the primary radical formed by ionization of the disulfide 

 compounds but that the healing of the molecule through capture of an electron 

 or later rupture of the charged link, probably by attraction of other groups, or 

 molecules, may occur so rapidly that this charged radical is not the one detected, 

 but rather the neutral radical R — S. However, our interpretations are still 

 tentative. Because we consider the question an important one we are continuing 

 to investigate it experimentally. Studies using S^^ can clear up this uncertainty. 

 What already seems established is that the odd electron giving rise to the pattern 

 is essentially localized on the sulfur. 



The large anisotropy in the g factor for the cystine-type resonance suggests 

 the potential usefulness of this resonance for obtaining structural information 

 about the proteins. Studies by Shields and the author on single crystals of 

 cystine (19) showed a resonance simpler and much narrower than that for 

 polycrystalline cystine, and one which shifted position sensitively with orienta- 

 tion in the magnetic field. After this observation the same crystal was crushed 

 up and found to give the resonance pattern characteristic of polycrystalline 

 cystine, shown in Fig. 4. Observations (19) on strands of hair and on feather 

 quill at various orientations in the d.c. magnetic field shovved only the poly- 

 crystalline type of cystine resonance for all orientations. It is significant, we 

 think, that the cystine-like resonance in these proteins is not orientation- 

 dependent, for that fact gives convincing proof that the bonds to sulfur, either 

 the C — S or S — S links, in the keratins are randomly oriented (in contrast to 

 hydrogen bridges). We have also made measurements (19) on stretched and 

 unstretched hair and found no significant variation in its cystine-hke resonance 

 pattern. In all cases it is like that of the polycrystalline cystine. 



The resonance of x-irradiated insuhn may exliibit a third type of protein 

 resonance (cf. Fig. 12). It has the characteristic sulfur or cystine-like pattern 

 but with a relatively sharp resonance superimposed (at the left of the pattern). 

 Although it has the same g factor — that of the free electron spin — this com- 

 ponent to the left seems too sharp to be classified as an unresolved doublet 

 like that of feather quill or silk. Possibly this sharp component of the insulin 

 resonance may arise from an electron trapped in one of the unsaturated ringed 

 structures known to be on the side chains of this protein. The ringed structure 

 may act as a sink or trap for the odd electron produced by ionizing radiation. 

 We have found a similarly sharp resonance (29) for x-irradiated polystyrene, 

 where the odd electron observed is believed to be trapped in the aromatic rings 

 attached to the backbone structure. 



Considering the varied patterns which we found for the resonances of the 

 different amino acids and simple peptides, it was at first surprising to us that 



