COMMCTNICATIONS TO THE EDITOR
Oct. 20, 1957
5*579
specific hydroxylation of aromatic compounds.1 non-specific hydroxylation of aromatic substances, When salicylic acid is hydroxylated by this sys- oxyferroperoxidase appears to be involveda6 A complete description of these experiments will tem, 2,3-dihydroxybenzoic and gentisic acids are formed in equal amounts, together with smaller be published elsewhere. quantities of trihydroxybenzoic acids. We have (6) This study has been supported by a grant from the United carried out this reaction in an atmosphere of 0ls2,States Public Health Service (A-971). (7) Post-doctorate Fellow of the United States Public Health in HeO solution free of heavy metal contaminants according to the method of Keilin and Hartree2 Service. H. S. MASOX and have separated the products on silica gel col- DEPARTMEXT OF BIOCHEMISTRY I. ONOPRIENKO umns. Oxygen which was incorporated into sali- UXIVERSITYOF OREGONMEDICAL SCHOOL K. YASUNORU OREGON D. BUHLER? cylic acid as hydroxyl t o form 2,3-dihydroxyben- PORTLAND, zoic and gentisic acids was found to arise entirely RECEIVED AUGUST12. 1957 from the atmosphere. When the reaction was carried out in H201*with an atmosphere of 02,no excess of labelled oxygen was detected (Table I). Ar,-3 PARTICIPATION I N THE BICYCLO [2,2,1]HEPTANE SERIES
Sir:
TABLE I SOURCEOF OXYGENINCORPORATED AS HYDROXYL IXTO SALICYLIC ACIDBY THE SYSTEM: HORSERADISH PEROXIDASE,^ DIHYDROXYFUMARATE,'
OXYGEX'
The rearrangement of the bicyclo [2,2,1]heptane carbon skeleton, proceeding via carbonium ions, is a well known phenomenon.' We wish to report the observation of a novel, base-catalyzed analog of this type of transformation. (OCOCH,
1, 2
018z
+ H2016
3
O1Sz
+ H201*
2,3-Dihydroxybenzoic acid Gentisic acid 2,3-Dihydroxybenzoic acid Gentisic acid
106 110 0
0
We gratefully acknowledge a gift of purified horseradish peroxidase (RZ = 1.9) from Professor D. Keilin. * Recrystallized from mixtures of acetone and deionized distilled water; Emol.= 8350, 308 mp, in ether. c Prepared by electrolysis of H20 containing 1.4 atom 7 0 0l8(Stewart Oxygen Company, San Frarcisco). Hydroxylations were carried out in 20 ml. solutions containing phosphate (0.046 dd), salicylic acid (300 pmoles), peroxidase (1.2 pmoles) and dihydroxyfumarate (0.81 mmole), brought to pH 6.0 with heavy-metal-free KOH. The mixture was acidified after two hours of reaction, and the products extracted with ethyl acetate. These were separated on silica gel columns, using chloroform and 4% ethanolic chloroform developers. Identities of the recovered gentisic acid (m.p. 202-203") and 2,3-dihydroxybenzoic acid (m.p. 205-206') were verified both spectroscopically and chromatographically. Under the conditions described, the combined yield (based upon a quantitative chromatographic procedure) was 16-21 %. With heat-denatured peroxidase, a yield of 1.9% was observed. e We are grateful t o Dr. C. C. Delwiche for performing the mass spectrometry w o n samples which we obtained by Unterzaucher pyrolysis3 of the hydroxylation products. An apparent isotope effect shown by our results is under study.
Since complex I11 of peroxidase predominates in the dihydroxyfumarate-oxygen system4 and the enzyme is inhibited by carbon monoxide, displaying the spectrum of carbon monoxide ferroperoxidase, 4 ~ s since dihydroxyfumarate (for which peroxidase is an oxidase) cannot be replaced by ascorbate in the hydroxylating system,' and since in this system molecular oxygen is activated toward (1) H . S. Mason, I. Onoprienko and D. Buhler, Biochim. Biophys. Acta, 24, 225 (1957). (2) D. Keilin and E. F. Hartree. Biochem. J . , 6 0 , 310 (1955). (3) W. E. Doering and E. Dorfman, THISJOURNAL, 75, 5595 (1953). (4) H.Theorell and B. Swedin, Nafurwiss.,27,95 (1939); B. Swedin and H. Theorell, Nature, 146,71 (1940). (5) B. Chance, J. B i d . Chem., 197, 577 (1952).
OR
The key compound in this study is the epoxide IIA (m.p. 107.5-108.0°; calcd. for CIF,HI~O~:C, 65.69; H , 5.15. Found: C, 65.97; H, 5.16), prepared by treatment of IA2 with peracetic or perfluoroperacetic acid. IIA reacted rapidly with hot 2 N sodium hydroxide to give an unstable phenolic product, the infrared spectrum of which lacked the strong 11.7 p (epoxide) band of IIA. Acetylation of this material with pyridine-acetic anhydride gave a product from which the tetraacetate IIIA (m.p. 139-140'; calcd. for C19H2008; C, 60.63; H, 5.36; acetyl, 46.7. Found: C, 60.47; H, 5.27; acetyl, 45.64) could be isolated in 20'% yield. An authentic sample of IIIA was readily obtained by acid hydrolysis of TIA, which on the basis of the behavior of norbornene epoxide3 would be expected to give the rearranged product, followed by acetylation. An attractive interpretation of this reaction involves nucleophilic opening of the epoxide ring by an internal phenoxide ion as shown below. Further attack by base on the dienone intermediate would regenerate the aromatic ring and give rise to a rearranged product.
A close analogy for this series of reactions is provided by the recent realization by Winstein and Baird of Arl-3 phenoxide participation in the (1) See, for example, C. K. Ingold, "Structure and Mechanism i n Organic Chemistry," Cornell University Press, Ithaca, N. Y., 1953, pp. 482-494. (2) W. R.Vanghan and M. Yoshimini, J . Org. Chcm., 23, 7 (1957). (3) H.M. Walborsky and D. F. Loncrini, THIS JOURNAL, 76, 5396 (1954); H.Kwart and W. G. Vosburgh. i b i d . , 76,5400 (1954).
5550
Yol. 79
COXI>IUNIC.~TIONSTO THE EDITOR
methanolysis of 2-p-hydroxyphenyl-1-ethyl broi ~ ~ i d eI .n~this instance, the intermediate dienone has actually been characterized.j I n support of these suggestions, it should be noted that norbornene epoxide is extremely inert toward basic hpdrolysk6 The special role of the internal anion is demonstrated in one further way. Lithium aluminum hydride reduction of IA gave the hydroquinone, I B (1n.p. 1-14.0L144.5’ ; calcd. for C11Hlo02; C, 75.84; H, 5.79. Found: C , 1, which upon treatment with diand base gave the dimethyl ether IC (m.p. 78.5-79.0‘; calcd. for C13H1402: C, -11.20; H, 6.98. Found: C ? 77.31; H , 6.71). Peracetic acid converted IC into the epoxide IIC (m.p. 119.5-120.0°; calcd. for C13H1403: C ,71.54; H, 6.47. Found: C, 71.56; H, 0.09). This epoxide was unreactive toward either sodium hydroxide or methanolic sodium methoxide under conditions which resulted in opening of the epoxide ring i n 11-1.7
0r CH,6
(i;
that the products of this condensation are HSIG C o 9 and Co4SH according to reaction 1. CHBCO-S-CoXI H20
+ CH~COCH~CO-S-CC):I” +
--+ CH~COHCR~CO-S-CO.I~’
f
I
CH2COOH Co:\-SH’
(11
The enzyme was assayed by measuring the disappearance of the enolate ion absorption of Xc,ic Co-1 a t 310 mp5 which occurs when Ac COX is added in the presence of the enzyme, P-Keto-thiolase,6 which interfered with this assay, was completely and irreversibly inhibited by treatment with concentrations of iodoacetamide which only partially inhibit the condensing enzyme. During the course of the reaction, for each equivalent of AcXc CoA which disappeared one equivalent of free thiol appeared as measured by the nitroprusside reaction7 (Table I). When acetylI
T.4BLE
The reaction was run in Beckman cuvettes, d = 1.0 cm Each contained 15 mg. of bovine albumin, 400 pmoles of Tris-HC1 buffer pH 7.75, 2.0 pmoles of AcCoA a n d 1.0 pinole of AcAc CoA. The final volume was 3.0 ml. Expt. 1 contained 1.8 mg. of enzyme protein, and Expt. 2 contained 0.6 mg. of enzyme protein, The incubation period was 2 0 minutes a t room temperature.
The partial support of this research by a research grant from the Kational Institutes of Health is gratefully acknowledged.
AAcAc CoA, pmoles
ASH, pmoles
-0,54
$0.58 +0 . 69
Expt. 1 Expt. 3
-0.68
l-CI4 CoA was incubated with AcAcCoA and the
(4) S. Winstein and R. Baird, THIS J O U R N A L , 79, 75G (19,;7) enzyme, one equivalent of XcAc CoA disappeared ( 5 ) R. Baird and S. Winstein, ibid., 79,1230 (1957). for each equivalent of Aa CoA which was incor(6) Unpublished observation of Norman Hudak. (7) Acid-catalyzed opening of epoxide IIC, followed by hydrolysis, porated into HXlG CoA as determined by radiogave t h e rearranged diol i ( m . p . 117.3-148.0°; calcd. for C I I H ~ C O ~ : activity in HMG (Table 11). C, 6 6 . 0 8 ; H, 6.83. Found: C , G S 9 2 ; €1, 6.92), xhose structure is substantiated by its failure t o reduce periodic acid in t h e standard vicinal diol test. This observation indirectly support? etructure 111 for the analogous product derived from acetolysis of I I A . (8) On leave from Kybto University, KyBto. l a p a n . (9) Opportunity Fellow, John H a y 1Vhitney Foundation. 1!)331956; Allied Chemical and Dye Corp Fellov, 1%G-1937.
TABLE I1 Each Beckman cuvette contained 15 mg. of bovine alburniu, 400 pmoles of Tris-HC1 buffer pH 8.1, 2.4 nig. of enzyme protein, 0.4 pmole of XcAcCoA and 1.0 pmolc of : i r e t ~ - i - l - CCo.4 ~ ~ with a specific activity of 97,5,000 counts] rnin.,/pmole. The incubation period was 20 min. a t room DEPARTMEST OF CHEXISTRI’ JERROLD >xEISTALD temperature. Radioactivity of H M G was determined by CORXELL CSIVERSITY HITOSIN O Z A K I ~the method of Rudney.4 GEORGE A. KILEY~ ITHACA, SEW YORK Total radioAmount of activity found Ac (20.4 inRECEIVED AUGLXT28, 1957 AAcAc CoA, in HRIG, corporated, THE BIOSYNTHESIS OF 3-HYDROXY-.jMETHYLGLUTARYL COENZYME A ‘
Expt. 1 Expt. 3
pmoles
counts/min.
pmoles
-0.18 -0.19
180,000 170,000
0.185 0.173
The product of the reaction with acetyl-1-C” Co-1 as reactant was treated with neutral hyPrevious work with rat liver and yeast prepara- droxylamine to convert the COX esters to hytions has established that -1zcC0~1~ and BcAlcCoA1 droxamates, and chromatographeds after the addiare the reactants in the biosynthesis of HhIG by tion of mono HIIG- and acetohydroxamates as carthe HXlG condensing e n ~ y n i e . ~In, ~ this coin- riers. Only two radioactive spots were observed, munication we wish to report the results of experi- corresponding to acetohydroxamate and HMGments with a purified preparation of condensing hydroxamate. XcAc Co-1 did not form a detectenzyme from baker’s yeast, which demonstrated able hydroxamate under these conditions6 These results show that the H X G formed during the re(1) This investigation was .upported in part by grants from t h e action is in the form of an acyl derivative of COX. Life Insurance Medical Research F u n d , T h e American Cancer Society, and t h e Elizabeth Severance Prentiss F u n d of Western Reserve The above experiments demonstrate the stoichiUniversity. T h e C‘4 used was obtained on crllocation from the Atomic ometry of reaction 1. Energy Commission.
Sir:
( 2 ) T h e folloning abbreviations are used: h c C0.4 and AcSc CoA, acetyl and acetoacetyl coenzyme A ; H I I G , 3-hydroxy-@-methylglutaric acid: H U G CoA, 0-hydroxy4 niethylglutaryl coenzyme A ; CoilSH, reduced coenzyme X . (3) H . Rudney, F c d e i n f i o i i Proc., 1 5 , 342 (1936). (4) H. Rudney, . I . B i d . C h t w . , 227, in pres- (1957).
( 5 1 J . R . Stern, i b i d . , 221, 33 (1056). J . R . Stern, If, J. Coon and -4.del Campillu, i b i d . , 2 2 1 , 1 (1950). ( 7 ) R. R . Grunert and P. H . Phillips, A r c h . Biochetn., 3 0 , 217
(e)
(195lj.
18) A. R . Thompson, A i ~ s i J. . S c i . Res., B4, 180 (1951).
