Recent Developments in the X-Ray Study of Haemoglobin 



one pair of haems is associated with each half. The two pairs, there- 

 fore, must be related by the diad axis and they must lie on opposite 

 sides of the plane of splitting. (3) The haem groups can be pictured 

 as disks of 15 A diameter and 3-7 A thickness. These must be so 

 placed that they fit into the unit cell of the dried as well as the wet 

 crystals. For instance, if the haem group stuck out above and below 

 the top and bottom surfaces of the globin cylinder, there would not 

 be room for them in the dried unit cell ; hence they are more likely 

 to be attached to the sides of the cylinder. (4) An azide group can be 

 attached to each iron atom without causing a change in unit cell 

 dimensions. Since an azide group is over 4 A long, this means that 

 the haem groups cannot be concerned in the bonding of neighbouring 

 haemoglobin molecules in the crystal lattice. 



This is as far as the data go at the moment. We may hope that 

 circumstantial evidence will continue to accumulate and that by 

 fitting this together bit by bit we may eventually be able to define 

 the positions of the haem groups within close limits. 



COMPARATIVE STUDIES 



Derivatives of Horse Haemoglobin — Having spent so much time on 

 an intensive x-ray analysis of methaemoglobin of horse I was naturally 

 curious to see how this structure compared with those of other 

 haemoglobin derivatives of the same species, especially whether the 

 physiologically more important derivatives oxy-, carboxy- and reduced 

 haemoglobin have the same molecular structure as methaemoglobin. 

 I soon found that the situation was complicated by the polymorphism 

 of many of these substances. It seems that met-, oxy- and carboxy- 

 haemoglobin can be crystallized in either of two forms. One is the 

 monoclinic one described in the foregoing pages and the other an 

 orthorhombic one, with a structure far less favourable for analysis 

 (mentioned in /, Table I, p. 93). Crystalline oxy- or carboxyhaemo- 

 globin can be changed into methaemoglobin without change in crystal 

 structure and therefore any structural conclusions derived from an 

 analysis of the latter apply with equal force to the former. In addition 

 to the two forms just mentioned, carboxyhaemoglobin has recently 

 been crystallized in a second monoclinic form, entirely different from 

 the one described here and containing a more complex arrangement 

 of the molecules in the unit cell. 



As F. Haurowitz discovered by optical studies 6 , the crystal structure 

 of reduced haemoglobin is different. Crystals of methaemoglobin 

 cannot be transformed into reduced haemoglobin without being 

 broken up, and the isotropic, pseudo-hexagonal plates of reduced 



145 



H— 10 



