54 



CHARLES TANFORD 



[3 



0-2301- 



0-220 



0-210 



0-200 



0-190 



0-180 



0-170 



0-160 



0-150 



0-140 



0-130 



0-120 



0-110 



J L 



1 I I I 1 1 1 



12 3 4 5 6 



7 

 pH 



8 9 10 11 12 13 14 



Fig. 10. Fluorescence polarization of ovalbumin (curve A) and of serum albumin (curve 

 C) (Weberio5). 



hydrogen bonds (or ion pairs) between side chain groups contribute to the 

 stabiHty of a compact configuration. (Since both the acid and alkaline reac- 

 tion are reversible, we are dealing in this case with true thermodynamic 

 stabihty.) However, an alternative explanation is possible. It is not neces- 

 sarily true that serum albumin, or any other protein, has a unique compact 

 configuration. More than one such configuration may exist. That at the iso- 

 electric point and acid to it may have a relatively high electrostatic interaction 

 energy. On the alkaUne side, where some of the charge-bearing side chains 

 are different from sites which are charged on the acid side, it may be possible 

 for the molecule to rearrange itself to an alternate compact configuration of 

 lower electrostatic free energy. (A transition from model A of Fig. 2 to 

 model B would be an example of the kind of transition here suggested.) 

 As we shall mention later, it has in fact often been suggested, for reasons 

 quite unconnected with the present discussion, that serum albumin has the 

 ability to undergo configurational changes of this kind. 



Hemoglobin. Hemoglobin undergoes a major configurational change, char- 

 acterized by a change in absorption spectrum and in the titration curve, 

 near pH 3-5. '^-^^ The reaction is relatively slow, becoming much faster with 

 decreasing pH, and is slowly reversible. The reaction is accompanied by 

 an increase in intrinsic viscosity (at pH 3-5 and ionic strength 0-04) from 

 3-5 to 13-5 cc./gram.^^ This is presumptive evidence for a randomly-coiled 

 configuration, though the effect of charge and ionic strength must be deter- 

 mined to establish this conclusion. The titration curve^^ on the alkaUne 



