HUGO THEORELL 



ically different. In cytochrome c, both the iron-Unked groups 

 are Hberated simultaneously, with jfcK' 2.12 and n = 2 (1). 

 This means that the liberation of one of the groups, by the 

 addition of a proton to the nitrogen atom, favors the disruption 

 of the other bond, too, perhaps by changing the shape of the 

 molecule. Chloride ions in acid solution give an interesting 



\+ / 

 NH 



N^cr N 



NH 

 ^ \ 



compound, "Cyt-2H+2C1~," presumably by entering between 

 the positively charged iron and nitrogen atoms. The magnetic 

 measurements revealed the iron to have 3 unpaired electrons, 

 and thus to be held by square dsp^ bonds, essentially covalent in 

 the plane of the heme disc, but more ionic in the perpendicular 

 direction. 



The hemopeptide was studied in the Spinco ultracentrifuge 

 at/?H 2.3, 4 and 8.7. AtpH 8.7 it was found to be polymerized 

 to a penta- or hexamer, whereas it was a monomer at pH 2.3. 

 This finding explains the occurrence of hemochromogen spectral 

 bands in alkaline solution ; the iron may be coupled on one side 

 to the imidazole of its own peptide, on the other side to the 

 terminal valine-NHo, or the e-amino group of lysine, of another 

 peptide, so that a five- or six-membered ring of hemopeptides 

 is formed. The spectral bands of the reduced hemopeptide 

 polymer were a little different from those of ferrocytochrome ; 

 however, when free histidine was added, the absorption bands 

 became spectrophotometrically indistinguishable from those 

 of ferrocytochrome c, and furthermore, depolymerization to 

 monomers was observed in the ultracentrifuge. These results 

 seem to speak strongly in favor of our earlier assumption that 

 imidazole is attached to the iron atom on both sides in the 

 native cytochrome c. 



282 



