629 



Rufus Lumry 



number of years ago Takashima was able to show that large changes in di- 

 electric relaxation time and dipole moment occurred on oxygenation of hemo- 

 globin providing the experiments were carried out at very low salt concentra- 

 tions. ^' At physiological salt concentrations no such effects could be obser- 

 ved in viscometry experiments. (^^' ^3) Muirhead and Perutz' ' have recently 

 found on comparison of horse deoxyhemoglobin and human oxyhemoglobin 

 crystals obtained from high-salt solutions that there is no detectable change in 

 internal folding of the four quarters of hemoglobin, but that the P sub -proteins 

 which form two of these quarters are positioned quite differently in the two 

 proteins. If, as is thought, it is possible to consider this difference a true 

 measure of the effect of oxygenation, we can conclude that the change in state 

 of iron produced by oxygen binding results in significant alteration in quaten- 

 ary structure. However, it is most innportant to note that this reorganization 

 must also result from changes in internal folding since there is no other way to 

 couple the state of the iron complex to the factors determining quatenary bind- 

 ing. ^^■^' The resolution of the x-ray diffraction method for hemoglobin is only 

 about 5 angstrom units so considerable variation in internal folding can occur 

 without detection. There is little reason at present to believe that changes of 

 much more than one or two angstrom units in some atom positions are re- 

 quired and the changes in geonrietry are detectable in the molecule as a whole 

 only because the delicate balance of interactions among the quarters produces 

 in the quatenary changes an amplification of the small internal changes. It is 

 interesting to note that what may be all-important small shifts in the positions 

 of many atoms during physiological reactions may not generally be detectable 

 by x-ray nnethods even at highest resolution. It may be hoped that this is too 

 pessimistic a view. The results of Takashima suggest that the largest changes 

 in protein conformation occur at intermediate states of oxygenation rather than 

 on formation of fully oxygenated hemoglobin. It may prove possible to crystal- 

 lize these intermediates for x-ray studies. 



The small shifts within the quarters of hemoglobin appear to be concentrated 

 in the region of the fifth -position imidazole ligand behind the heme plane. 

 Evidence largely from Antonini and Wyman and their co-workers^ ' has 

 drawn special attention to this part of the protein and, taken with the fifth -posi- 

 tion imidazole linkage between iron and protein already discussed, presents 

 reasonable qualitative explanations for most of the long -puzzling peculiarities 

 of hemoglobin reactions. Only one of these need be discussed here. This is 

 the so-called heme -heme interaction in which the replacement by oxygen of 

 water at the sixth ligand position of the iron ion of one sub-protein increases 

 the affinity for oxygen on the iron ion of a second sub -protein. At present this 

 picture is best explained by the following steps: (1) oxygenation of one iron ion 

 causes conformational readjustments in the region of its fifth-position ligand; 

 (2) through the strong coupling between the sub -proteins these dislocations are 

 spread to a second sub-protein to cause a readjustment of the fifth-ligand imi- 

 dazole group of that sub -protein; (3) the change in the orientation of this ligand 

 alters the electronic properties of the second iron ion to increase its affinity 

 for oxygen. Thus the chemical reactions of two distinct sub -proteins are 

 coupled by interactions through "'the conformations of the sub-proteins. The 

 situation is actually more complicated than this since there are dependencies 

 on the total structure of the protein. The total protein must have a very rigid 

 conformation and we have begun to suspect that the high content of a -helix in 

 these proteins is needed to establish rigidity. The a -helix is a very rigid 

 structure' ' and undoubtedly stiffens the protein much as iron bars will stiffen 

 a rubbery matrix. Even cytochrome c has a high helix content which we can 

 currently estimate as forty per cent. 



