April, 1951
ENZYMATIC EVIDENCE FOR DESOXYRIBONUCLEOTIDE STRUCTURES [CONTRIBUTION FROM THE
BIOLOGY DIVISION,O A K RIDGENATIONAL
LABORATORY
1537
]
Enzymatic Evidence for the Structure of Desoxyribonucleotides BY C. E. CARTER Desoxyribonucleotides prepared by enzymatic degradation of thymus desoxyribonucleic acid and isolated by ion-exchange chromatography are dephosphorylated by an enzyme isolated from bull semen which hydrolyzes the phosphate group of nucleotides esterified in the 5’-position and is inactive against ribonucleotides derived from ribonucleic acids. Desoxyadenylic acid is deaminated by muscle adenylic acid deaminase. The phosphate group of desoxyadenylic acid is labile in acid and alkali, whereas the phosphate group of adenosine-5’-phosphate is relatively resistant to hydrolysis. These data are interpreted to support a 5’-nucleotide structure of desoxynucleotides.
The position of the phosphoester linkage in desoxynucleotides and makes comparative rate desoxyribonucleotides has been considered analo- studies of hydrolysis of the phosphoester groups gous to that demonstrated for several ribonucleo- with the analogous ribonucleotides questionable tides derived from ribonucleic acid, namely, the evidence for structure. Because of the well-defined specificity of the 5’third carbon of the carbohydrate moiety of the molecule.’ Although the finding of isomeric ribo- nucleotidase and muscle adenylic acid deaminase* nucleotides raises the question of Cs-phosphoester the finding that desoxynucleotides are substrates linkages in these corn pound^,^^^ which as yet re- for the enzymes supports a 5‘-phosphoester strucmains unsettled, similar considerations do not ture for these compounds, a conclusion which apply to desoxyribonucleotides because of the must remain tentative until proof of structure by absence of the Cz-hydroxyl. Furthermore, there chemical degradation and synthesis is established. is no evidence for isomerism in the desoxyriboExperimental nucleotide^.^ The position of the phosphoester linkage must then be either CBor CS. Evidence The 5’-nucleotidase employed in these studies was genfor the structure of the desoxyribonucleotides iso- erously supplied by L. A. Heppel of the National Institutes of Health. Specificity studies conducted in this Laboralated by ion-exchange procedures4 has, in the tory c o n k n those published by Heppel.’ The preparation present investigation, been obtained from experi- and characterization of the desoxyribonucleotides is dements employing these compounds as substrates scribed by Volkin, Khym and Cohn.‘ Muscle adenylic acid deaminase was prepared by first for a specific 5‘-nucleotidase purified from bull extracting myosin from rabbit muscle by the procedure desemen by H e p ~ e l (and , ~ ~also ~ ~Mann6) and, in the scribed by Szent-Gyorgy,O separating this fraction from the case of desoxyadenylic acid, muscle adenylic acid insoluble proteins by pressing through cheese cloth and then deaminase. It has been found that desoxyaden- extracting the deaminase from the latter portion by homoylic, -guanylic, -cytidylic and - thymidylic acids as genizing for 30 minutes in 4 volumes of 0.1 N ammonium Following centrifugation in the cold, a turbid well as muscle adenylic acid are all dephosphoryla- acetate. supernatant solution was obtained which contains most of ted by the 5‘-nucleotidase, whereas riboadenylic, the original muscle deaminase activity. -guanylic, -cytidylic and -uridylic acids are not Enzymatic Assay with 5’-Nucleotidase.-Substrates were substrates for the enzyme. Desoxyadenylic acid dissolved in PH 6.2 veronal buffer to give a concentration of 100 pg. of nucleotide phosphorus ml. The enzyme is deaminated by a muscle adenylic acid deaminase was dissolved in distilled water to give a concentration of a t a rate slower than that for adenosine-5’-phos- 0.5 mg. per ml. For enzymatic assay 0.1 ml. of enzyme phate but to complete deamination. was incubated a t 38’ with 0.25 ml. of substrate and 0.05 The hydrolysis of the phosphate group of the ml. of 0.01 M magnesium chloride solution. Tubes were a t intervals and inorganic phosphorus determined desoxynucleotides in acid and alkali has been removed by the Fiske-SubbaRow method. All desoxyribonucleostudied and it was found that the phosphoester tides were completely dephosphorylated by the enzyme but groups of desoxyadenylic and desoxyguanylic acid a t different rates as shown in Fig. 1. None of the monoare hydrolyzed in 1 N hydrochloric acid a t approxi- nucleotides derived from ribonucleic acid were degraded the enzyme. Muscle adenylic acid (adenosine-5‘-phosmately the same rate as riboadenylic and ribo- by phate) and desoxyadenylic acid were enzymatically deguanylic acids. In 1 N sodium hydroxide inor- phosphorylated at the same rate. ganic phosphate is liberated more rapidly from Enzymatic Assay with Muscle Adenylic Acid Deaminase. desoxyadenylic acid than from muscle adenylic -Adenosined ’-phosphate and desoxyadenylic acid were dissolved in 0.1 N succinate buffer PH 6.0, to give a conacid or yeast adenylic acid (“A” and ‘iB’’).2*a centration of 15 pg. of nucleotide per ml. To 3 ml. of these It is probable that the chemical lability of desoxy- solutions, 0.1 ml. of enzyme solution was added andothe diribose7 as contrasted with ribose accounts for the gest incubated in spectrophotometer cuvettes a t 26 . The hydrolytic behavior of the phosphoester in the course of the enzymatic deamination was followed spectro(1) € Brederek, I . “Fortschritte der Chemie organischer Naturstoffe,” Band I, Julius Springer, Berlin, 1948, p. 121. (2) C. E. Carter, THIS JOURNAL, 71, 1466 (1950). (3) W. E. Cohn, ibid., 73, 1480 (1950). (4) E. Volkin, J. X. Khym and W. E. Cohn, ibid., 78, 1533 (1951). (5a) In a personal communication Heppel has reported that this enzyme hydrolyzes nicotinamide ribose-5’-phosphate and synthetic uridine-5’-phosphate and cytidine-5’-phosphate as well as adenodneand inosine-5’-phosphate. (5) L. A. Heppel, Federation Proc., 9, 184 (1950). (6) J. T. Mann, Biochcm. J . , 89, 346 (1946). (7)W. R. Tipson, “Chemistry of Nucleic Adds in Advance in Cuhohydrate Chemistry,” Vol. 1. Audemic Rem, New York,N.Y . , 194b, p. 198.
photometrically by Kalckar’s method.6 Aslshown in Fig. 2 , both adenosine-5’-phosphate and desoxyadenylic acid are deaminated by the muscle enzyme, the rate for the latter substrate being slower although both reactions go to complete deamination. Yeast adenylic acids (adenylic “A” and “B”)a,* are not substrates for the enzyme. The products of the enzymatic reactions were identified by paper chromatography. Acid and Alkaline Hydrolysis.-The desoxynucleotides were adjusted to a concentration of 25 pg. of nucleotide phosphorus in 1 ml. of 1 N hydrochloric acid and placed in (8) H.M. K d c h t , J . Bid. Chem., 167, 4413 (1947). (9) A. Scent-Oyorgy, “Muaculu Contraction,” Aclrdemic P r u , New York, N.Y..1946, p. 186.
C. E. CARTER
1538
Vol. 73
IO0
lo[ 90 (80 O l’
300-
I
2501 200
E“
g
0-0
0
A/
150-
N
a w I
100-
/A
/’
050-
/’
A/A‘I
---- d----
/ -+--j
T
0
IO
20
AA’
2’
*+*---30
AJA
40
50
, 60
70
20
I
40
60
April, 1951
DESOXY-5-METHYIXYTIDYLIC
a reflection of the general lability of purine desoxyribosyl compounds rather than an indication of the position of esterification of phosphate on the desoxyribose residue for, as shown in Fig. 4, the phosphoester linkage of desoxy-
[CONTRIBUTIONPROM
THE
ACID FROM THYMUS NUCLEICACID
1539
adenylic acid is more labile in 1 N sodium hydroxide than yeast or muscle adenylic acids.
6, OHIO CLEVELAND
RECEIVED AUGUST28, 1950
BIOLOGY DIVISION,OAK RIDGE NATIONAL LABORATORY]
The Isolation and Identification of Desoxy-5-methylcytidylic Acid from Thymus Nucleic Acid’ BY WALDOE. COHN Ion-exchange fractionation of enzyniatic digests of DNA from thymus has resulted in the isolation of a desoxynucleotide of 5-methylcytosine in amounts small compared to the amount of cytidylic acid present. This conlirms the earlier reports on the isolation of 5-methylcytosine from acid digests of DNA and suggests that this base occurs in desoxyribonucleic acids as a nucleotide similar to the others, being released upon enzymatic digestion. new nucleotide present in the original desoxycytidylic acid. Introduction The desired fraction was concentrated by reabsorption and Since the time that Johnson and Coghill re- elution with 0.1 M formic acid. All subsequent tests were portedza the isolation, from an acid hydrolysate of carried out upon this formic acid solution which contained tuberculinic acid, of a substance crystallographi- approximately 2 mg. per ml. Phosphate’ and desoxypendeterminations8 upon the nucleotide and spectrophocally identical with synthetic 5-methylcyto~ine~~~ it tose tometric comparison of the derived base with synthetic 5has been an open question as to whether this sub- methylcytosine, kindly supplied by Dr. Hitching9 of Wellstance was a bona fide constituent of nucleic acids come Laboratories, were carried out in this Laboratory, as in the usual combination with a pentose phosphate. were all ion-exchange tests. Paper chromatographic of the nucleotide, the derived base, and the enJohnson and H a r k i d demonstrated that none of analyses zymatically deaminated1@derivative of the base (thymine), this material could be isolated from “yeast nucleic as well as spectrophotometric comparison of both bases with acid.” Aside from this observation, there was synthetic substances, were most generously made by Drs. no follow-up of the original finding; it remained an Erwin Chargaff and Jacob Kream of Columbia University. uncorrelated oddity until very recently when Results Hotchkiss o b s e r ~ e din , ~ the base mixtures derived (a) Identification of the Base.-Hydrolysis of from desoxyribonucleic acid preparations, a small the nucleotide was accomplished by heating for amount of a cytosine-like base which he termed 40 minutes on a steam-bath in 72% perchloric “epicytosine.” Although it seemed likely to him acid.loa The digest was diluted tenfold with water that this substance was 5-methylcytosine, the and poured through a 1 cm. X 1sq. cm. bed of sulamounts available for study were too small to per- fonic-acid type cation-exchanger to recover the mit identification. pyrimidine base which was then eluted with amIn the course of isolating gram amounts of the de- monium hydroxide. This substance had the specsoxyribonucleotides from enzymatically degraded trophotometric properties, compared to synthetic DNAJ6the presence of a hitherto unobserved sub- 5-methylcyto~ine,~ shown in Fig. 1. It could not stance was detected in the desoxycytidylic acid be absorbed upon a strong base anion-exchanger exfraction.6 From approximately 800 mg. of the cept in the hydroxy form, which is characteristic of crude desoxycytidylic acid, enough of this sub- cytosine alone among other nucleic-acid bases. l1 stance (ca. 20 mg.) was isolated to permit identifiHydrolysis12in 99% formic acid a t 175’ for 2 cation of it as the desoxypentosyl phosphate of 5- hours followed by paper chromatographic analysisla methylcytosine. The isolation and identification in butanol-water for 16 hours gave but one compoof this nucleotide are described in this communica- nent detectable by ultraviolet fluorescence. l 4 This tion. component lay in the same position as 5-methylExperimental cytosine (RF 0.28), which differs from cytosine The source of 5-methylcytidylic acid was the desoxycyti- (0.22), uracil (0.36) and thymine (0.52). When dylic acid fraction prepared from thymus DNA by Volkin, eluted from the paper, it had a spectrum identical Khym and Cohn.6 The separation was performed on the same ion-exchange column (33 sq. cm. X 12 cm. 200-500 with that of synthetic 5-methylcytosine. Another mesh strong-base anion exchanger, in formate form) with portion of the formic acid hydrolysate, containing 0.003 M formic acid. The 5-methylcytidylic acid fraction 159 pg. of the pyrimidine, was incubated with puriwas reabsorbed on a smaller column and refractionated to remove the last traces of desoxycytidylic acid; this was necessary because of the very low amount (ca. 3%) of the
(1) Work performed under Contract W-7406-eng-28 for the Atomic Energy Commission. (2) (a) T. B. Johnson and R. D. Coghill, THIS JOURNAL, 47, 2888 (1925); (b) H. L. Wheeler and T. B. Johnson, A m . Chcm. J . , 31, 591 (1904); cf. ref. 10. (3) T. B. Johnson and H. H. Harkins, THISJOURNAL, 61, 1779 (1929). la R. D. Hotchkiss, J . Bid. Chem.. 176, 315 (1948). (5) E. Volkin, ‘J.!X.Khym and W. E. Cohn, THISJOURNAL, 71, 1633 (1951). (6) W.E. ‘Ahd,
