lO J. C. KENDREW 



definite possibility that it is glutamic acid. It is probable that we shall have to wait 

 for a more highly resolved Fourier synthesis before this question can be answered 

 definitely. 



There are indeed serious problems about the attachment of large ligand groups. 

 We have found that /)-iodophenyl isocyanide can be diffused into a crystal of myo- 

 globin and that it combines at the haem group, producing the characteristic change 

 in spectrum. When one looks at the model of the myoglobin molecule, and notes 

 how closely the side chains are packed together, it is hard to understand how a 

 ligand as large as this can approach the iron atom without a major disturbance in 

 the structure. It may indeed be that some disturbance does take place, because we 

 find that the /)-iodophenyl /^ocyanide derivative of myoglobin crystallizes with 

 slightly different cell dimensions from the normal crystal, and on some occasions 

 assumes a totally new crystal form, which suggests that the overall shape of the 

 molecule may have been slightly changed. In this connexion we are contemplating 

 the possibility of making comparative studies of the detailed structure of myoglobin 

 with diflFerent ligands attached to the haem group. 



Theorell: Of course, this change in shape of haemoglobin molecule on oxygen- 

 ation is very interesting indeed, as it has been since the observation over 30 years 

 ago of the strictly hyperbolic oxygenation curve of the myoglobin. Do you think 

 the S shape could be a consequence of the change in shape of the molecule ? Could 

 it explain why the introduction of the first oxygen for instance is so diflficult and 

 the latter ones so much easier ? 



Kendrew : I agree that it is quite possible to imagine that the first oxygen 

 becoming attached to haemoglobin in some way alters the relative position of the 

 sub-units, so that subsequent oxygens can enter more readily. This idea, however, 

 is purely speculative at present. 



Theorell: It would be very interesting to know if under low oxygen tension 

 the whole change occurs at the introduction of the first oxygen molecule. 



Kendrew: It would be very difficult to study this question experimentally 

 unless somebody could discover a method of preparing crystals of haemoglobin in 

 the partly oxygenated state. 



Hirs: In applying the method of isomorphous replacement to myoglobin you 

 prepared several heavy atom derivatives with reagents such as mercury diammine, 

 aurichloride, etc., in which you were subsequently able to locate the position of the 

 heavy atoms in regard to the 6 A resolution structure. Can you now tell us with 

 which residues these derivatives are formed ? I believe the information would be of 

 interest to some of us. 



Kendrew : The heavy atom groups which we used for working out the structure 

 of myoglobin undoubtedly cause local disturbances on the side chains in their 

 immediate neighbourhood. These disturbances mean that it is particularly difficult 

 to identify just those side chains which are of most interest in the present context. 

 In succeeding stages of the analysis we propose to use conventional refinement 

 methods which do not involve the introduction of heavy atoms at all ; if these 

 methods are successful the problem of local disturbances will not arise and we shall 

 then be in a position to identify the important side chains. Our impression is that in 

 most cases the heavy atom groups do not combine in a strictly chemical fashion 



