96 Discussion 



the similarities, there are also important differences between these two classes of 

 compounds. 



I wonder if his evidence against the imidazole hypothesis actually reflects an over- 

 looked structural difference between oxy- or carboxyhaemoglobin and the other 

 complexes. Recent studies by Y. N. Chiu and myself on the pH-dependence of the 

 reaction between carbon monoxide and haem adsorbed on poly-1-vinylpyrrolidone 

 showed a pronounced Bohr effect when histidine was used as the sixth ligand, but no 

 apparent Bohr effect when pyridine or water was used as the sixth ligand as expected 

 from the Pauling-Coryell hypothesis. 



Models for Linked Ionizations in Haemoproteins 

 By P. George, G. I. H. Hanania, D. H. Irvine and N. Wade (Philadelphia) 



George : With reference to linked ionizations in haemoproteins, it is remarkable how few 

 data there are for simple co-ordination compounds possessing an ionizing group in 

 the vicinity of the co-ordination centre, that can serve as a guide to the change in 

 pAT that might be expected when such a ligand is bonded to a metal. 



Hanania and Irvine {Nature, Lond. 183, 40, 1959) have recently carried out an in- 

 vestigation of pyridine-2-aldoxime and its ferrous complex, rather similar to that 

 reported by Williams on Fe(DMG)2(imid)i^, and have obtained complete thermo- 

 dynamic data for the ionization of the oxime OH-group in both the free ligand and 

 the complex. 



Consideration of the structure of the ferrous complexes shows how resonance 

 stabilization would favour the formation of the anion, and an increase in acid strength 

 would therefore be expected; but no quantitative predictions can be made. The free 

 oxime has ipK = 10-2 at 25°C and I = 0: ^H° = 6-0 kcal/mole and A5° = -26 e.u. 

 In the Fe++-tris-pyridine-2-aldoxime complex the ionization of the third oxime group 

 occurs with a pK of about 7-0 so that in both cases the charge change is from zero to 

 — 1 , and the ionizations are therefore strictly comparable. For the ionization of the 

 group in the complex AH° = 1-4 kcal/mole and AS° = —27 e.u. Hence, on co- 

 ordination, the acid strength of the oxime OH-group is increased by about 3 pH units, 

 corresponding to a favourable change in the free energy of ionization of 4-3 kcal/mole. 

 It is particularly interesting that this increase is borne almost entirely by the change 

 in AH°: as can be seen the entropies of ionization are nearly identical. 



Turning now to haem-linked ionizations in haemoproteins, Hanania, Wade and I 

 have recently been studying the pH variation of the equilibrium constant for the 

 formation of the imidazole complex of ferrimyoglobin. Russell and Pauling (Proc. 

 nat. Acad. Sci. 25, 517, 1939) published a few results on the ferrihaemoglobin deriva- 

 tive, and reported a pK of about 9-5 for the NH ionization. From an analysis of our 

 data we find the p^ for the ferrimyoglobin complex to be about 10-0 at 25°C. The 

 ionization can be represented as follows: 



Prot - FcMb^ - ImHv^ Prot - FeMi3+ - Im- + H+ 



where ImH stands for the neutral imidazole molecule and Im~ for the anion. From 

 spectrophotometric measurements in the region 225 to 250 m/i, we have found a pAT 

 value lying between 13 and 13-5 for the corresponding ionization in the free ligand; 

 i.e. 



ImHv^ Im- + H+ 



Hence, on co-ordination of the glyoxalinium N-atom to the Fe of ferrimyoglobin, 

 the acid strength of the imino NH group increases by about 3 to 3-5 pH units. This 

 change is very similar to that found by Hanania and Irvine {loc. cit.) for pyridine-2- 

 aldoxime and its ferrous complex, and we are extending the measurements to obtain 

 the AH° and A5° values. 



The \)K values for the haem-linked ionization in ferrimyoglobin derivatives lie 

 between 6 and 7 (George and Hanania, Disc. Faraday Soc. 20, 216, 1956). Thus, 

 if the group responsible is to be identified as the imino NH group of a histidine residue. 



