Sept. 5; 1958
COMMUNICATIONS TO THE EDITOR
4751
toration of activity. Addition of zinc in excess of and 697 (ebullioscopic in butanone, acetic acid, .,75 mp, 1 gram atom per mole does not result in additional ethanol, respectively) ; m.p. 49.9'; X2 activity and may inhibit; this excess zinc can be 405 mp, ethanol (oxidized form) and Xmax.290 mp, removed by dialysis a t neutral @H. Similar results ethanol (reduced form) z. By reductive acetylation are obtained when the zinc is removed a t PH 4.2, in the diacetate of dihydro I (m.p. 40') (11) was prethe presence or absence of OP, and the protein pared (Anal. Found: C, 79.41; H, 9.71; 0-Me, 6.50, 7.08; acetyl, 9.96; moles HZ absorbed per further treated as above. Dialysis of the metal-free protein against ions of 100 g., 1.05; 1000 [X-ray], 1100 (isothermal Group IIA and IIB elements--hlg++, Ca++, distillation)). Levulinic acid, acetic acid and SUCcinic acid have been identified as products of alkaCd++, and Hg++-did not restore activity. On the other hand, dialysis of the metal-free line potassium permanganate treatment of I. =1 paraquinone function is indicated by the spectra protein against ions of the first transitional periodof I and 11. The other two oxygens are accounted Cr+++, >In++, Fe++, Co++, and Ni++-resulted for by two alkoxyl groups. A polyisoprenoid side in a significant restoration of activity. The present experiments indicate the crucial nature chain(s) with one double bond per unit is indicated of the metal atom as anactive component of the enzy- by the oxidation products and hydrogenation. matic site of this exopeptidase. Carboxypeptidase X-Ray analysisg shows a pseudo-periodicity which can be isolated with its full complement of zinc, thus is expected for such a side chain. The consumption differing from many metal activated systems. This of 11.1 moles of hydrogen, based on a molecular zinc can be successively removed and then be re- weight of 863,1° indicates ten mono-unsaturated placed. Therefore, carboxypeptidase constitutes a isoprenoid units. This work has been confirmed and extended in suitable model system to explore the significant difference in the physical-chemical and enzymatic proper- accompanying communications. ties of metalloenzymes and metal-enzyme-complexe~.~ We also have succeeded in isolating from miThis work was supported by grants-in-aid from the crobial sources fourI2 other crystalline compounds National Institutes of Health of the Department of which we believe to be homologs of I, differing from each other in the length of their polyisoprenoid Health and Welfare. side chains. On the basis of the formula C59(6) B. L. Vallee, Advances in Protein Chem., 10, 317 (1955). Hgo04" for I, these compounds are formulated as BIOPHYSICS RESEARCH LABORATORY OF THE DEPARTMENT OF MEDICINE BERTL. VALLEE 111, Cj4H8204, Torula yeast, m.p. 45.2'; IV, HARVARD MEDICAL SCHOOL AND THE JOHN A. RUPLEY C4gH7404, A . vinelandii, m.p. 37"; 17, C44H6604, PETER BENTBRIGHAM HOSPITAL THOMAS L. COOMBS BOSTON,MASSACHUSETTS HANSNEURATH Torula yeast, m.p. 30.5'; VI, C39H~804,S. cerevisiae, m.p. 16". All data available support these DEPARTMENT OF BIOCHEMISTRY UNIVERSITY OF WASHINGTON conclusions. The ultraviolet and visible absorpSEATTLE, WASHINGTON tion spectra of all these compounds (I, 111, IV, V, RECEIVED AUGUST6, 1958 VI) are qualitatively identical, indicating a common quinonoid chromophoric group; equivalent weight COENZYME Q : A NEW GROUP OF QUINONES' (alkoxyl, oxygen, oxidation-reduction titration of Sir: hydroquinone) and molecular weight determinations The isolation of crystalline quinone (Q-275) (I) (X-ray) confirm that these compounds differfrom the from lipides of beef heart mitochondria has been next lower homologue by a five-carbon isoprenoid d e s ~ r i b e d . ~ I, ~acts . ~ as a coenzyme by undergoing unit. I n particular, cell dimensionsgof I , 111,IV subcyclic oxidation and reduction during substrate stantiate this point. The estimate of the number of oxidation in mitochondria.2Jt6 Evidence bearing isoprenoid groups based on hydrogenation data on the possible role of I in oxidative phosphory- would be 9.4,8.0,7.0 for 111,IV and VI respectively. lation also has been pre~ented.'.~ Anal. Found : Since all five quinones have coenzymatic acC , 82.24 rt 0.32 (8); H, 10.38 A 0.27 (8); 0, tivity,6 we have altered our designation Q-275 to (Unterzaucher) 7.60, 7.71; 0-Me, (Zeisel) 6.78 coenzyme Q to represent this new group of com=t 0.16 (6); C-Me, (Kuhn-Roth) 13.4 =k 2.1 (3); pounds. On the basis of the apparent number of moles Hz absorbed per 100 g., 1.29; acetyl, zero; isoprenoid units, we have designated the individual equiv. wt., 447; mol. wt., 900 and 910 (X-ray dif- members of the coenzyme Q group as coenzyme fractiong), 779 (isothermal distillation), 527, 632, Qo l (I); coenzyme Qg (111); coenzyme QE (IV); coenzyme Q7 (V) , coenzyme 9 6 (VI). (1) This investigation mas supported in part by grants H-458(C8), Ubiquinone (SA) was described by Morton, 2G&?(C8), and R6-5506, National Institutes of Health, U. S. Public Health Service. et al.13-16as preparations (m.p. range 33-41") from
(2) F. L. Crane, Y.Hatefi, R. L. Lester and C. Widmer, Biochim. Biophys. Acla, 26, 220 (1957). (3) F. L. Crane, R. L. Lester, C. Widmer and Y . Hatefi, ibid., in press. (4) R. L. Lester and F. L. Crane, ibid., in press. (5) F. L. Crane, C. Widmer, R. L. Lester and Y. Hatefi, ibid., in press. (6) Y.Hatefi, R. L., Lester, F. L. Crane and C. Widmer, i b i d . , in press. (7) Y. Hatefi, R . L. Lester and T. Ramasarma, F e d . Proc., 17, 238 (1958). (8) Y. Hatefi, Biocnim. Biophys. Acta, in press. (9) We are greatly indebted to Dr. Caroline MacGillavry of the University of Amsterdam for the X-ray determinations and their interpretation.
(10) D. E. Wolf, C. H. Hoffman, N. R. Trenner, B. H. Arison, C. H. Shunk, B. 0. Linn, J. F. McPherson and K. Folkers, THIS JOURNAL, 80, 4752 (1958). (11) C. H. Shunk, B. 0. Linn, E. L. Wong, P. E. Wittreich, F. M. Robinson, and K. Folkers, i b i d . , 80, 4753 (1958). (12) R. L. Lester, Y.Hatefi, C. Widmer and F. L. Crane, Biochim. Biophys. Acta. in press. (13) G. N. Festenstein, F. W. Heaton, J. S. Lowe and R. A. Morton, Biochcm. J . , 69, 558 (1955). (14) 1. C. Cain and R. A. Morton, ibid., 60,274 (1955). (15) F. W. Heaton, J. S. Lowe and R. A. Morton, J . Chem. SOL., 4094 (1956). (16) R. A. Morton, G.M. Wilson, J. S. Lowe and W. M. F.Lent, Chem. b I n d . , 1649 (1957).
4752
COMMUNICATIONS TO THE
EDITOR
i'ol. YO
animal organs and yeast (oily fractions). I t is clearly related to our coenzyme Q group. The authors are indebted to Dr. David E. Green for his encouragement in the course of these investigations.
27.5 and 407 in@)identified ; i s %,3-diniethoxy-3,Gdime thylbenzoquino~ie~. \Ye have studied the nuclear magnetic resonance spectra a t 40 mc. of Qlo and many synthetic model compounds, aiid Dr. James N. Shovlery (Varian CONTBIBUTION FROM R . I,. LESrEB Associates) has kindly determined the spectrum a t 00 me. These data characterize the INSTITUTE F O R ENZYME RESEARCH F. L CRANE of these data siio\i. convincingly that THEUNIVERSITY OF \#ISCOXSIN Y 114TEFI protons of QO: MADISON, WISCONSIN Ql0hss t'it o CH30--, one CH,-- and one isoprenoid RECEIVED A u c v s ~13, 1058 chain nf 10 units attached to a benzoquinone nucleus, but do not define the position of the ring --substittients. S.in.r. data esclude the presence of COENZYME Q. I. STRUCTURE STUDIES ON THE I COENZYME Q GROUP aromatic pi oh^^, -C=CH2 and --C&C&, and Sir: assign the ten double bonds in the isoprenoid side The discovery of a quinone, 9-27;, has been chain as in I. 'The n.rn.r. spectruin of eicosahyreported, and characterizing chemical and physical drocoenzyme Qlb siio\%sthe one =CCE& nucleus properties, oxidative degradation and hydro- group a t - 111..5 C.P.S. While this group is not genation are reported in an accompanying com- clearly resolved from -C-CH_,CI~,C== in the munication.2 Four closely related quinones having spectrum of Qlna t 40 mc., i t is a t GO mc. Reduction and methylation of Ql0 with dimethyl similsr coenzymatic activity have been isolated from microbial sources.2 AI1 five new quinones are sulfate, gavs a colorless crystalline tetramethoxy designated as members of a coenzyme Q group, derivative (IV), m.p. 38-30", [ a ] D 0' (chloroform). Foucd: C,81.85; H, 10.82,0CHd,14.2. ?.e., coenzyme Q6,( 2 7 , Q, Q g and QIO. Oxidation of IV with about a 4-foId e x e s s of We noTx have extended these observations. Formula I agrees with our structural data for aqueous alkaline permanganate a t 100' yielded coenzyme QIo (beefheart) and I1 and 111correspond tetran;etho:iS.pht~ialic anhydride (V) (after sublimation), imp. 13Y-1395 not depressed by admixto Q9,Qe, etc. ture of synthetic tctramethoxyphthalic anhydride.
Our isolated Qlo (yellow) melted a t 49.5-50.5'. Found: C, 81.98, 82.05; IH, 10.38, 10.31. H y drogenation of Qlo resulted in an absorption of about 11 moles of hydrogen. Oxidation of the resulting hydroquinone yielded eicosahydro-coenzyme Qlo; ),Laooctsne 278 mp, 187. Found: C, S0.30; H, l?%6. h'o.. of ~. ~
Proton type
HC= C&O=C-C&CH= =c-Cgr4!&--c= CI3C-( nucleus) CHsC=(chain)
i
C.P.S.~
+8 -34 -64, -69 -113
-125
protons No. of calcd. protons for Rela- based CrsHootivc on 0 4 and band 2CHsO/ strucareas mole ture I
5 3 1 20
10 0 2 40
16.5
33
-
10
6 0
Y
(36 (3 33
-
'.'
,.. I.
Oxidation of I V in acetone with the stoichiometric quantity of permanganate for ten double bonds gave aftcr partition chromatography, 2methyl-3,4,5,6-tetramethoxyphenylacetic acid (VI) m.p. 75-76', identical by melting point behavior and infrared spectra with a synthetic sample. The presence of levulinic and succinic acids as oxidation products was confirmed.2 The chemical and physical properties of synthetic 2,3-dimethoxy-5-methyl-6-farnesylbenzoquinone4 and analogs' are closely similar to those of Qlo. Ubiquinone! reported by Morton, et al., from animal organs (n1.p. range 33-41°)Gand yeast (yellow oily fractions)G is clearly related to the coenzyme Q group. DONALD E. WOLF CARLH. HOFFMAS NELSONR. TRENNER BYRON H. ARISON CLIFFORD H. SHVNK BRUCE0. LIXN JAMES F. MCPHBRSON KARLFOLKBRS RECEIVED A r ~ r 13, ~ s1958 ~
91 90 The bands refer to 40 mc. spectra in carbon tetrachloride, means a t lower fields than water protons while - means a t higher fields.
CONTRIBUTION FROM THE MERCK,SHARP& DOI.IMB RESEARCH LABORATORIES DIVISIONO F MERCR & eo., ISC. NEWJERSEY RAHWAY,
Coenzyme Q1o appears to be a 2,3-dimethoxybenzoquinone derivative. I t s absorption spectrum with maxima a t 275 mp and 405 mp is in good agreement with that of aurantiogliocladin, (maxima a t
-
+
7.
(1) P. L. Crane, Y. Hatefi, K. 1.. Lester and C. Widmer, Bidchim. Aiophw. Ac:o, 25, 220 (19571. (2) B. L. Lester, F. L. Crane and Y. Hatefi, T H I S J O U R N A L , 80,
(8) E. B. Visber. J . CAW:. SOL., 815 (1953). (4) C. H. Shunk. B. 0. I.inn. E. L. Wong, P. E. Wittreicb. I?. 37. Robinson and K. Folkers, Txirs J O I X N A I . . 80, 4753 (1958). ( 5 ) R. 8. Morton, G. If,Wilson, J. S. Lowe and W. M. I;. Lent, Chcm. & I n d . , 1649 (1957). ( 6 ) F. W. Henton, J S Lowe and R. A. Morton. J . Chem. SOC.,
4751 (1968).
4094 (19BG)
