The Isolation, Purification and Properties of Haemin a 



331 



of native haemoglobin a, and decreasing the pH to below 9 shifts the absorption 

 band of the alkali-denatured globin-compound from 575 to 596 m/t. The reaction is 

 also given by chlorocruorohaem and monoformyldeuterohaem. The absorption 

 spectrum of porphyrin a is not altered by the addition of denatured globin to its 

 alkaline solution. The reaction is not caused by alkali alone; pyridine haemochrome 

 a is not altered by alkali; nor is the band at 573-5 m/< produced by imidazole, NH3, 

 glycylglycine, aminoacids (histidine, lysine, tyrosine) or SH-compounds (cysteine, 

 GSH, thioglycolate) in 0-1 N NaOH. Possibly a Schiflf's base is formed between the 

 formyl groups of the haems and amino groups in the proteins, resembling alkaline 

 'indicator yellow', but having the instability to lower pH characteristic of retinyl- 

 idenemethylamine (Morton and Pitt, Biochem. J. 59, 128, 1955). 

 Williams: I wish to show why the model systems of Lemberg and co-workers do not 

 reproduce the physical properties of cytochrome a^. The predicted change in the 



, High spin type 





.^ Low spin type 



X Y 



Basicity of ligond 



Fig. 1. 



Soret band position of ferrous porphyrin complexes with change in the basicity of 

 the ligand in the fifth and sixth positions is illustrated in Fig. 1. 



For completely ionic high-spin complexes, 0-X, the band position is predicted to 

 move to longer wavelengths with increase in basicity. The same shift is expected for 

 completely covalent complexes, Y-Z. In the region X-Y z. change of magnetic 

 moment occurs and there is a chemical equilibrium involving two spin states. In this 

 region the Amax falls with the increase of basicity of the ligand. Complexes of small 

 molecules such as CO, O2, CN~ cannot be compared with other ligands by their 

 basicity. They shift the absorption bands to longer wavelengths for reasons given in 

 my paper. The point, Q, represents a typical band position for these low-spin com- 

 plexes. Now if we have an ionic complex giving a band at long Amax, then the addition 

 of CO will move the band to shorter wavelengths. A complex giving a band at position 

 Y will show a band shift to the red on addition of CO. The models of Lemberg fall 

 in the general region just before Y. Cytochrome 03 and to a slightly lesser degree 

 cytochrome a fall near to X. Cytochromes Z> and c fall beyond Y with the other 

 haemochromes of protoporphyrin and mesoporphyrin. I consider that nearly all the 

 observations of Lemberg and co-workers are consistent with this interpretation. 



The argument can also be extended to the visible region of the spectrum. Here the 

 plot of Amax (-x and /3 bands) against ligand basicity is only slightly different (Fig. 3, p. 48 

 of this volume). In the ionic complexes there is little shift of /Imax with basicity and the 

 greatest change in band position is found in the region X- Y. The work of Falk (this 

 volume, p. 74) defines the region beyond Y. 



We observe from the two diagrams that in complexes with ligands of very low 

 basicity the Soret band will move in the opposite direction to the a and /J bands on 

 the addition of carbon monoxide. This is observed with FeP(H202). Turning to 

 Lemberg's data on the visible spectra we find that whereas cytochrome a is very 



H.E. — VOL. I — X 



