772
NOTES
Acknowledgments. The author is grateful to Dr. D. E. Soule and Dr. F. G. Young for discussions. The assistance of Miss L. I. Forrest and C. W. Nesbeda in carrying out the experimental work is appreciated.
The Radiation-Induced Oxidation of Peptides in Aqueous Solutions' by Harriette L. Atkins, Winifred Bennett-Corniea, and Warren M. Garrison Lawrence Radiation Laboratory, University of California, Berkelev, California (Receiaed August 1.2, 1966)
Radiolysis of certain compounds of the type RCONHCHR2 in dilute aqueous solution containing O2 leads to formation of labile peptide derivatives which are readily degraded on mild hydrolysis to give ammonia and carbonyl compounds as characteristic products. It has been proposed2v3that such oxidation in the case of peptides derived from the simpler amino acids, glycine and alanine, is initiated by a preferential attack of the OH radical at the carbon-hydrogen position CY to the nitrogen function. The over-all reaction scheme includes the radiation-induced step4 (1) H20 -+ H202, H2, OH, H, esq-, H+ followed by
OH
RC0NHCHR2
0 2
+ RCO"CR2
----f H2°
(2) (3)
RCoNHcR2
RCO"C(OZ)RZ
+
The reducing species, eaq- and H, are scavenged quantitatively by 0 2 to form 0 2 - and HOz, which are related H+. Proposed by the equilibrium6 HO2 a 02reaction modes for removal of the intermediate peroxy radicals include
+
RCONHC(02)Rz +RCON=CRz HOz
+ RCONHC(02)R2
+ HO2
(4)
+
0 2
(5)
+ HzO2
(6)
--j
RCONHC(OOH)R2 HzO
+ RCONHC(OOH)R2
-+
RCONHC(0H)h
where RCON=CR2 represents a dehydropeptide and RCONHC(OH)R2 the corresponding hydrate. Such compounds readily decompose on hydrolysis6 RCON=CR2
+ 2H20 + RCOOH
The Journol of P h y a i d Chemistry
+ NH3 + RzCO
(7)
RCONHC(OH)R2
+ HzO
-+
RCOOH
+ NH3 + R2CO
(8) The above reaction scheme requires that the ammonia and carbonyl yields be in the relationship G(NH3) = G(R2CO) = GOHwhere the latter term represents the 100-ev yield for OH production in the radiation-induced step 1. We have measured ammonia and carbonyl yields in the y-ray-induced oxidation of Nacetylgly cine,3c glycine anhydride, 3d N-acet ylalanine, and alanine anhydride in oxygenated 0.1 M solution and we find-for each system G(NH3) 'v 3, which value is consistent with recent measurements' of the maximal yield of OH radical in the y radiolysis of water containing reactive solutes. On the other hand, we also find that carbonyl production in these simple peptide systems is not in accord with the quantitative requirements of the proposed oxidation scheme; the initial carbonyl yields are uniformly low with G(R2CO) I 0.8. There is then the question as to whether this apparent discrepancy arises from (a) an incorrect formulation of the locus of initial OH attack or from (b) unspecified complexities in the chemistry of removal of the peroxy radicals RCONHC(02)Rz. To obtain specific information on this point, we have employed Fe(II1) instead of O2 as the scavenger of intermediate radicals formed in the radiolysis of Nacetylalanine and N-acetylglycine. Heavy metal ions such as Fe(II1) and Cu(I1) oxidize organic free radicals in aqueous solution by electron transfer and by ligand transfere8 Such reactions in the case of the peptide radical RCONHcR2 would correspond to RCONHCR~+ Fe(II1) + RCON=CR2
R C O N H ~ R 2+ Fe(III)
+
H20
RCONHC(OH)R2
-
+ Fe(I1) + H +
+ Fe(I1) + H +
(9) (10)
(1) This work was performed under the auspices of the U. S. Atomic Energy Commission, AEC Contract No. W-7405-eng-48. (2) M. E. Jayko, B. M. Weeks, and W. M. Garrison, Proc. Intern. Conf. Peaceful Uses A t . Energy, Geneva, )nd, 29, 30 (1958). (3) {a) W. M. Garrison, M. E. Jayko, and W. Bennett-Corniea, Radzation Res., 16,483 (1962); (b) W. M. Garrison and B. M. Weeks, ibid., 17, 341 (1962); ( c ) W. M. Garrison, Radiation Rea., Suppl., 4, 148'(1964); (d) W. Bennett-Corniea and W. M. Garrison, UCRL8988, Dec 1959. (4) (a) A. 0. Allen, Radiation Res., Suppl., 4, 54 (1964); (b) E. J. Hart, Science, 146, 19 (1964). (5) G. Csepaki and H. J. Bielski, J . Phys. Chem., 6 7 , 2180 (1963). (6) J. P. Greenstein and M. Winits, "Chemistry of the Amino Acids," John Wiley and Sons, Inc., New York, N. Y., 1961, p 843. (7) E. Hayon, Trans. Faradall Soc., 61, 723 (1965). (8) J. H. Baxendale and D. H. Smithies, 2. Phgsik. Chem., 7 , 242 (1956); H. E. DeLaMare, J. K. Kochi, and F. F. Rust, J. Am. Chem. SOC.,8 5 , 1437 (1963).
NOTES
773
respectively. We note that the organic products of reactions 9 and 10 are identical with the postulated products of reactions 4 and 6.
,
I
I
I
I
I
Experimental Section Solutions containing the acetylamino acid (Nutritional Biochemicals, twice recrystallized) plus ferric sulfate (CP) were adjusted to pH 3 with sulfuric acid to retain Fe(II1) in solution, evacuated, and then irradiated with Co6"y rays to a dose of not more than -5 X 10l8 ev/ml. Above this value the dose-yield curves were no longer linear. Dose was determined with the Fricke dosimeter, G(Fe(II1)) = 15.5, C F ~ ( I I I ) 2130 at 305 mp and 22". Immediately after irradiation, an aliquot, was passed through a Dowex-50 column (1 cm X 10 cm), acid form, to remove iron. The eluent was made 2 N in hydrochloric acid, heated to 100" for 60 min, cooled, and assayed for a-keto acid and aldehyde through use of the 2,4-dinitrophenylhydrazine reagent.s A second aliquot was made 2 N in sodium hydroxide, allowed to stand 24 hr in the outer compart,ment of a Conway diffusion cell to liberate ammonia which was collected in 0.1 N sulfuric acid in the inner compartment, and assayed by the standard Tollen procedure. Control runs established that product hydrolysis and ammonia transfer were quantitative under these experimental conditions.
0 0
Fern
eaq-
H
+ Fe(II1) -+
Fe(I1)
(11) (12)
+ Fe(II1) +Fe(II) + H+
and under these conditions the yield for peptide oxidation via reaction 2 followed by reactions 9 and 10 is in accord with -G(peptide)
=
G(NH3) = G(RC0COOH)
N
3.2
+ GH*o%
GOH
0.15
(E)
Figure 1. Effect of Fe(II1)concentration on the yields of ammonia ( 0 )and pyruvic acid (A)from 0.1 M acetylalanine and of ammonia (0) and glyoxylic acid ( A ) from 0.1 M acetylglycine.
I
I
I
I
I
I
I
I
t
Results and Discussion Ammonia production (Figure 1) in 0.1 M acetylalanine increases abruptly from G(NH3) = 0.7 to G (NH3) = 4.3 with increasing concentrations of Fe(II1) up to -low3 114. The ammonia yield then falls gradually to a limiting value of G(NH3) = 3.3 at the higher Fe(II1) concentrations. We also find under these conditions that pyruvic acid and ammonia are formed in equal molar yields. Yields of glyoxylic acid and ammonia from acetylglycine show this same quantitative relationship. Aldehyde yields from these systems are low, G 'v 0.1. At the higher (Fe(III))/(peptide) ratios of Figure 1, the reducing species, eaq- and H, are preferentially scavenged by Fe(II1)
0. IO
0.05
0
0.5 1.0 Acetylalanine ( E 1
1.5
Figure 2. Effect of acetylalanine concentration on yields of ammonia ( 0 )and of pyruvic acid (A)from solutions containing 0.05 M Fe(II1).
Hydrogen peroxide formed in the radiation-induced step 1 reacts rapidly with Fe(I1) to give an additional yield of OH radicals.'" The maximum in the yield curve shown in Figure 1 is attributed to the onset of the reaction H
+ RCONHCHRz -+ RCONHCRz + Ha
(13)
in competition with reaction 12 as the Fe(II1) concentration is reduced below 0.05 M . The RCONHCRz radicals from both reactions 2 and 13 are then available for oxidation by Fe(II1) via steps 9 and 10. The H (9) G. R. A. Johnson and G . Scholes, Analyst, 79, 217 (1954); T. E. Friedemann and G. E. Haugen, J . BioZ. Chem., 147, 415 (1943). (10) E'. Haber and J. Weiss, Proc. Roy. SOC.(London), A147, 333 (1934).
Volume 71, Number 3 February 1967
COMMUNICATIONS TO THE EDITOR
774
atoms involved in reaction 13 are derived from the radiation-induced step 1 and the conversion reaction eaq-
+ H+
---f
H
(14)
which at pH 3 is in competition with reaction 11 at the lower Fe(II1) concentrations. A t Fe(1II) concentrations below M , the dimerization of RCONHcR2 radicaWb becomes competitive with reactions 9 and 10 and the ammonia yield is depressed accordingly. If now the Fe(II1) concentration is kept constant at 0.05 M and the acetylalanine concentration is increased over the range lod3 to 1.5 M , it is found (Figure 2) that the ammonia and pyruvic acid yields are essentially equivalent and approach a limiting value of G 'v 3.9 at acetylalanine concentrations above 0.25 M . At these higher (peptide)/(Fe(III)) ratios, the H atoms formed in the radiation-induced step 1 are not scavenged by Fe(II1) but are preferentially removed via reaction 13. (The conversion reaction 14 at pH 3 is wholly quenched by 0.05 M Fe(III).) Hence we have the higher limiting oxidation yield which is in accord with the stoichiometry -G(peptide) = G(SH3) = G(R2CO) 'v
4.0 s!GOH
+ GH*o$+ GH
We conclude that the reaction of OH and H radicals with these peptide derivatives of the simpler amino acids, glycine and alanine, occurs essentially quanti-
tatively at the a position as formulated in reactions 2 and 3." The present data also establish the quantitative oxidation of RCONHCRz radicals by Fe(II1) via reactions 9 and 10. In the case of acetylalanine such oxidation appears to occur almost exclusively through ligand transfer (reaction 10) since measurements of the optical absorption of the irradiated solutions (after removal of Fe(II1)) reveal negligible absorption above 230 mp when read differentially against unirradiated control solution. Absorption by control solutions containing authentic acetyldehydroalanine12 (ezd0 6050) show that G values of >0.1 for reaction 9 would be detectable. To our knowledge the optical properties of acetyldehydroglycine have not been described. The low carbonyl yields obtained when O2 is used in place of Fe(II1) as the radical scavenger are interpreted here as evidence that other more complex degradation reactions occur in parallel with the dehydrogenation and hydroxylation reactions 4-6. The nature of these more complex branching reactions of the peroxy radical RCONHC(Oz)zis presently under study. (11) We refer t o radical attack a t the C-H linkage a to the nitrogen function as the characteristic peptide reaction. With the more complicated amino acid residues, attack may occur in parallel or exclusively at side-chain loci. See ref 3 and H. A.Sokol, W. BennettCorniea, and W. M. Garrison, J. Am. Chem. Soc., 87, 1391 (1965). (12) We are indebted t o Dr. B. M. Weeks for the sample of authentic acetyldehydroalanine which was prepared after the method of V. E, Price and J. P. Greenstein, Arch. Biochem., 10, 383 (1948).
C O M M U N I C A T I O N S T O THE E D I T O R The Absorption Spectrum of the Allyl Radical by Pulse Radiolysis' Sir: The identification of the optical absorption spectrum of the allyl radical has been the subject of much discussion and various investigators have suggested widely different values for its absorption maximaS2-' We have observed the absorption spectrum of the transient allyl radical in the pulse radiolysis of dilute solutions of allyl bromide in cyclohexane and have obtained a preliminary value for its absolute rate constant for termination. All of the pulse radiolysis experiments were performed using the linear electron accelerator at The Journal of Physicd Chemwtry
Argonne National Laboratory. A typical run consisted in introducing a several-joule pulse of 13.5-Mv electrons into a 15-ml sample at 22". Transient optical absorption analysis over a 16.0-cm optical path length was carried out in a manner similar to previous work.*l9 ~~~
~~~~
~~
~~
U.9. Atomic Energy Commission. (2) D. M. Bodily and M. Dole, J. Chem. Phy8., 44,2821 (1966). (3) M.B. Fallgatter and M. Dole, J . Phy8. Chem., 68,1988 (1964). (4)D. A. Ramsay, J. Chem. Phya., 43,818 (1965). ( 5 ) C.L. Currie and D. A. Ramsay, ibid., 45,488 (1966). (6)5. Ohnishi, S. Sugimoto, and I. Nitta, ibid., 39,2647 (1963). (7) H.C.Longuet-Higgins and J. A. Pople, Proc. Phy8. SOC.(London), A68,591(1956). (1) Based on work performed under the auspices of the
