BARBARA W. LOW AND JOHN T. EDSALL 



Although the portions of the serum albumin molecule are 

 tied together by the 17 disulfide bridges, the structure must 

 possess considerable flexibility (97), as shown by the virtually 

 instantaneous increase in optical rotation and in viscosity which 

 occurs when the molecule is exposed to high concentrations of 

 urea (45,54) or to acid pH values in the range 1.5 to 3.5 (111). 

 The changes are not only instantaneous but also largely reversible. 

 In contrast, ovalbumin, which has only one or two disulfide 

 linkages per molecule, shows slow changes in optical rotation 

 (101) and in viscosity (44) in urea solutions, which continue 

 over many hours. The native structure of ovalbumin appears 

 to be more rigid; it does not yield easily to stress, but once 

 opened up it is disrupted for good. Serum albumin, on the 

 other hand, yields readily to forces which tend to cause it to 

 unfold — either to the repulsive electrostatic forces produced 

 when the molecule acquires a large net charge or to the pene- 

 trating urea molecules which disrupt hydrogen bonds — but 

 returns to something closely approaching its original state when 

 the disturbing force is removed. We have already called 

 attention to a similar diflference between the two molecules with 

 respect to the ionization of their tyrosine hydroxyl groups. In 

 ovalbumin (28) these groups are firmly held by hydrogen bonds 

 and do not ionize until a. pH greater than 12 is reached, but the 

 breaking of the bonds, once achieved, is irreversible. In serum 

 albumin (103) some hydrogen bonds involving tyrosine appear 

 to exist, but they are readily broken in alkaline solutions and 

 the process is readily reversed on acidification. 



These differences, and many others which could be recorded, 

 indicate a profound difference in structural pattern between 

 these two familiar proteins. We have tried to suggest some of 

 the common recurring features, which appear to be of nearly 

 universal significance in protein architecture, and others which 

 may be related to the individuality — we may almost say the 

 caprice — displayed in the pattern of this or that particular 

 protein. Our vision of these molecular patterns is still, for the 

 most part, uncertain and dim. However, our understanding of 



428 



