258 



Discussion 



This presumably occurs through the two single equivalent steps, Fe"' -> Fe'^' followed 

 by Fe'^ -> Fe^'. Evidence that supports this is as follows. First, if the Fe"' derivative 

 is allowed to form by the spontaneous reduction of Fe^ , produced either by peroxides 

 or chloriridate, then the addition of chloriridate very rapidly effects the change 

 Fe"' -^ Fe^ . Secondly, if potassium molybdicyanide is used instead of chloriridate 

 as the one-equivalent oxidizing agent in the original reaction with ferriperoxidase, 

 only the Fe'^ derivative is formed. Then again, if chloriridate is added to the Fe'^ 

 derivative formed in this way, Fe'' results. Apparently chloriridate but not molybdi- 

 cyanide, under the experimental conditions employed, has a sufficiently high £■„' to 

 effect the oxidation of Fe'^ to Fe^ . Furthermore it is an interesting reflection on the 

 ability of peroxides to engage in net two-equivalent oxidations that they are com- 

 pletely ineffective in bringing about this second step, Fe'^ -^ Fe^ (George, Science, 

 117, 220, 1953; Currents in Biochemical Research, Ed. D. E. Green, 2, p. 338, 1956). 



Chance: The success of the experiment illustrated by the figure above stems from our 

 finding that the primary intermediate of bacterial catalase and methyl hydrogen 

 peroxide is sufficiently stable to allow its titration with the substrate (Chance and 

 Herbert, Biochem. J. 46, 4, 1950, p. 402). Although chemical depletion of peroxidase of 

 endogenous donor has given preparations in which the primary intermediate is more 

 stable (Chance, Arch. Biochem. Biophys. 41, 416, 1952), it is as yet inadequate for the 

 precision demanded of these titrations. 



The titrations with one-equivalent oxidants do suggest the mechanism outlined above 

 by George; however, it may be that interaction of the one-equivalent oxidant with 

 portions of the peroxidase protein could produce a two-equivalent oxidant — a possi- 

 bility that could not be disproved on stoichiometric grounds (Chance and Fergusson, 

 in The Mechanism of Enzyme Action, Johns Hopkins Press, Baltimore, 1954, p. 389; 

 Fergusson, /. Amer. chem. Soc. 78, 741, 1956). The experiments described here merely 

 afford a new approach to the problem posed by Eq. (1) and (2) above. It is, however, 

 a "kinetic" approach, and the interpretation of the result surely depends on the 

 reactions postulated. A more elegant test would be aftbrded by chemical determina- 

 tion of alcohol formation in Eq. (2), a topic on which active experimentation is 

 proceeding. 



DwYER : A simple system recently investigated by Craig, Dwyer and Glaser (Craig, Dwyer 

 and Glaser, /. Amer. chem. Soc, in press) may be useful to this discussion. Trimethyl- 

 amine N-oxide undergoes the rearrangement (CH3)3NO -^ (CH3)2NH -f- H-CHO in 



H,C 



I 



(CHJpN 



3'2 



Fig. 2. 



the presence of various metal complexes. The necessary conditions for the catalyst 

 complex containing Fe'", Ru'", or V'^' are (1) a site for the attachment of the N-oxide 

 through the oxygen, (2) an adjacent site containing OH, and (3) the complex must be 

 capable of oxidation. The intermediate complex shown in the figure is self-explanatory. 

 The metal atom, e.g. Fe is either oxidized or polarized to simulate Fe'^ by the oxygen 

 of the N-oxide. This oxygen then breaks the bond to the nitrogen carrying an electron 



