572 Discussion 



capable of oxidation to form aggregates, and (c) reactive with Cu++ ions. If thiol 

 groups of cysteine residues in the protein are bonded to the flavin by hydrogen bonding 

 (which would be influenced by changes of pH and ionic strength), this would provide 

 a structure which would meet all the known experimental requirements. We believe 

 that oxidation of the very reactive thiol groups by air suffices to explain the formation 

 of aggregates. 



Boundary analysis of the sedimentation of intact crystalline cytochrome b.^ shows 

 that the spreading of the boundary with time can be accounted for wholly on the basis 

 of diffusion, if this follows a Gaussian distribution (Baldwin, Biochem. J. 65, 503, 1957). 

 Extensive studies of cytochrome b^ and its modified forms by the approach to 

 equilibrium method for the ultracentrifuge determination of molecular weight lead to 

 the conclusion that under the conditions employed, cytochrome b^ may behave as an 

 interacting system, although nothing has been deduced as to the nature of the inter- 

 action. 



Margoliash: I should like to ask whether you have some evidence other than the reaction 

 of/j-chloromercuriphenylsulphonate for the idea that a sulphydryl group is an essential 

 binding point of the flavin to the protein in yeast cytochrome b,. /?-Chloromercuri- 

 phenylsulphonate is not strictly specific for — SH and could possibly react with other 

 amino acid side chains such as imidazole groups. Thiol groups in proteins also vary very 

 considerably in their reactivity to so-called sulphydryl reagents. Did you study the 

 kinetics and stoichiometry of the reaction either spectrophotometricajly or by esti- 

 mation of the protein-bound mercury ? Was an independent titration curve for the 

 — SH groups obtained ? N-ethylmaleimide is considered a more specific reagent for 

 — SH than the mercurials and would probably be safer in terms of interpretations of 

 the mechanism of liberation of flavin from cytochrome bo- 



Morton. The following evidence strongly suggests that thiol groups are involved in the 

 binding of riboflavin phosphate to the protein. 



1. Oxygen, Cu++ salts, HjOg and /j-chloromercuriphenylsulphonatc (PCMS) all 

 cause a loss of activity and associated fluorescence indicating a change in the bonding 

 of the flavin prosthetic group. 



2. Amino acid analyses and sulphur determinations suggest that there are about 

 eighteen i-cystine residues/haem. We have found that only 4 — SH groups/haem 

 react with PCMS, which completely displaces the flavin from the enzyme. 



We have not, as yet, studied the effect of N-ethyl maleimide on the enzyme. How- 

 ever, Boeri and colleagues showed that iodoacetate is a poor inhibitor of cytochrome 

 Z>2 as compared with PCMS. This suggests that some of the thiol groups essential for 

 activity are 'masked' by hydrogen bonding (or other type of bonding). Moreover, 

 the formation of only two aggregates, by oxidation, rather than a multiplicity of 

 aggregates, is evidence for the formation of — S — S — bonds at very specific sites of 

 the protein (see Armstrong, Coates and Morton, Nature, Lond. 186, 1033, 1960). 

 BoERi: In my experience the enzyme is sensitive to some inhibitors which act on ^SH 

 groups but not to others. The sensitivity to /?-chloromercuribenzoate is high. The 

 enzyme is less sensitive to o-iodosobenzoate. Monoiodoacetate does not inhibit at 

 the usual pH of the test, but is inhibitory at pH 5 or lower. The enzyme is very sensi- 

 tive at pH higher than 7 to sulphanilamide disulphides, as was shown by Brighenti. 

 These compounds act by a reaction of the type: 



Enzyme— SH -t- RSSR = Enzyme— SSR -f RSH 



Possible Free Radical Formation in Flavoproteins 



By C. D. LuDwiG (Philadelphia) 

 LuDWio: The comments which I have to make do not directly concern flavoprotein- 

 haem interactions, but pyridine nucleotide-flavoprotein interactions. However, they 

 might have some relevance, by analogy, to some of the findings of Morton, and Boeri 

 and their collaborators, in their studies of cytochrome bz- The work which I shall 

 describe was done in Theorell's laboratory by Ehrenberg and myself. 



