ACYL ADENYLATES: THE SYNTHESIS AND PROPERTIES OF ADENYL ACETATE* BY PAUL BERGt WITH THE TECHNICAL ABSISTANCE OF GEORGIA NEWTON (From the Department of Microbiology, Washington University School of Medicine, St. Louis, Misso~~ri) (Received for publication, March 9, 1956) A study of the mechanism of acetyl CoAl formation from ATP, acetate, and CoA by yeast aceto-CoA-kinase has revealed that the reaction occurs in two steps (1, 2). The evidence suggests that ATP and acetate react to form PP and the acetyl derivative of A5P, which is then utilized in the formation of acetyl CoA. This formulation is supported by the finding that synthetic adenyl acetate and PP are enzymatically converted to ATP (Reaction I), and that acetyl CoA is formed from adenyl acetate and CoA (Reaction 2). (1 ) ATP + acetnte e adenyl acetate + PP (2) Adenyl acetate + CoA F-: acetyl CoA + A5P The present communication deals with a description of the synthesis and properties of adenyl acetate. Materials and Jfelhods DPN, TPN, A5P, and glucose-6-phosphate dehydrogenase (0.8 unit per mg. (3)) were products of the Sigma Chemical Company, and CoA (75 per cent pure) was obtained from the Pabst Brewing Company. Hexokinase and P32P32 were prepared as previously described (2), and citrate-condensing enzyme (4) containing malic dehydrogenase was kindly supplied by Dr. S. Ochoa. A5P deaminase was prepared from rabbit muscle according to the method of Kalckar (5), and 5`-nucleotidase was obtained from bull semen according to Heppel and Hilmoe (6), by using the "simplified prepa- ration." * This work WLL~ supported by grants from the United States Public Health Serv- t Scholar in Cancer Resenrch of the American Cancer Society. 1 The following abbreviations have been used throughout. Acetyl coenzyme A, acetyl CoA; adenosine triphosphate, ATP; inorganic pyrophosphate, PP; adeno- sine-5'-phosphate, A5P; diphosphopyridine nucleotide, DPN; triphosphopyridine nucleotide, TPN ; trichIoronce tic acid, TCA ; tris (hydroxymethyl)aminomethane, Tris. 1015 ice and the National Science Foundation. lOlG .\DENYL ACEThTE SYNTHESIS Free A5P was determined by deamination with A5P deaminase (7) and by the liberation of inorganic phosphate (8) with 5'-nucleotridase (6), while acetate was measured with a partially purified preparation of acetokinase by the method of Rose et al. (9). Determination of Adenyl Acetule-Adenyl acetate was determined either colorimetrically as the ferric complex of acethydroxamate, or enzymatically by conversion to ATP or acetyl COX. In the first, adenyl acetate and P32P32 were incubated with the enzyme, and the ATP formed was determined after adsorptioii and elution from Norit (2). ATP formation was also determined by measurement of TPN reduction in the presence of aceto-CoA-kinase, hexokinase, glucose, glucose-6-phosphate dehydrogenase, and TPN (2). Acetyl CoA synthesis from adenyl acetate and CoA was measured by DPNH formation in the following series of reac- tions. ATP synthesis from adenyl acetatc was measured in two ways. (3) ace to-CoA-kinase Adenyl acetate + CoA + acetyI CoA + A5P + H+ malic dehydrogenase L-Malate + DPN t oxalncetate + DPNH + HC (4) citrate-condensing enzyme (5) Acetyl CoA + oxulacetate citrate + CoA + H+ (6) Adenyl acetate + IIPN + r,-malate + citrate + A5P + DPNH + 3H+ In the colorimetric procedure, 0.1 ml. of freshly neutralized 2 nr hydrosyl- amine was added to the sample in 0.9 nil., and after 5 minutes at 37", or at room temperature, 0.5 nil. of 10 per cent ferric chloride containing 0.2 K TCA and 0.66 N HC1 was added. After 5 minutes the optical density at 540 mp was determined in a cuvette with a light path of 1 cm. by using a Beckman DU spectrophotometer. A sample to which no adenyl acetate had been added served as the blank. 1 pmole gave an optical density of 0.630 under these conditions. There was good agreement between the values obtained in this manner and those determined enzymatically (Ta- ble I). During these experiments it was found that incubation of the ade- nyl acetate hydroxylamine mixture at 100" for 5 minutes, inst'ead of at 37", gave higher values for adenyl acetate than those found enzymatically. This discrepancy will be discussed later. Synthesis of Adenpl Acetate-Adenyl acetate was prepared in two ways. The first was the reaction of acetyl chloride and silver adenylate by a modi- fication of the method used by Lipmann and Tuttle for acetyl phosphate (lo), and the second was a modification of the method of Avison (11) with acetic anhydride and A5P in aqueous pyridine. P. BEltG 101'7 Acethydroxamate formation pnroles per id. 75.2 Procedure A-360 mg. of A5P were suspended in 20 ml. of water and the pH was adjusted to pH 6.5 with KOH. 4.0 ml. of 1.0 M silver nitrate were added and after 1 hour at 4" the precipitate was removed by centrifugation, washed twice with 25 ml. portions of cold water, twice with 50 ml. portions of cold ethanol, and twice with 50 ml. portions of ether. The precipitate was dried in vacuo over phosphorus pentoxide and paraffin. The yield of silver adenylate was 435 mg. 430 mg. of the silver adenylate were suspended in 5 ml. of anhydrous ether (from a freshly opened container) in a three-necked flask fitted with a dropping funnel and two drying tubes containing calcium chloride. The dropping funnel was also fitted with a drying tube at the top. The flask was immersed in an ice bath and the suspension stirred with a magnetic stirrer. 15 ml. of an ether solution containing 10 mg. of freshly distilled acetyl chloride per nil. were added dropwisc with vigorous stirring over a ATP formation Acetyl CoA formation &Inroles pn nrl. I #moles per ml. - ---I- ! 73.v I 74.9 TABLE I Enzymatic and Colorimetric Determination of Synthetic Adenyl Acetate * Measured by conversion of P3fP33 to ATP (2). t Measured spectrophotometrically by TPNH formation (2). period of about 20 minutes. The stirring ww'continued for an additional 15 minutes and then 5 ml. of cold water were added. The suspension was carefully adjusted to pH 6.5 by the addition of 0.2 M potassium carbonate, and then the ether layer was removed and washed twice with 5 ml. portions of cold water. The aqueous fractions were combined and the residual ether removed by blowing a stream of air over the solution at room temperature. To the above solution (30 ml.) were added 3.3 ml. of 1 N HC1 and, after centrifugation, the silver chloride precipitate was washed with 5 ml. of cold water and the wash and supernatant fluids were combined and carefully neutralized to pH 6.5 with KOH. This solution (38 nil.), based on its op- tical density at 260 nip (extinction coefficient 16 X lo3 em.-' M+ (7)), con- tained 20.3 pmoles of total A5P per ml. By measurement of acethydros- amic acid formation at 37", this solution contained 2.8 pmoles of labile acetyl groups per ml. Procedure B-2.0 gm. of A5P (5.81 millimoles) were suspended in 16 ml. of water and the pH was adjusted to about 7 with 8.5 bf KOH. Pyridine (4 ml.) was added and the solution was diluted to 24 ml. with water. The 1018 ADENYL ACETATE SYNTHESIS solution was cooled to -5", and acetic anhydride (3.4 ml., 35 mmoles) was added with vigorous stirring over a period of 3 minutes. 5 minutes later 375 ml. of acetone (-15') were added and after 15 minutes at -15" the solution was centrifuged. The precipitate formed at this stage was ge- latinous, and washing was difficult. Therefore, the precipitated material was dissolved in 10 ml. of cold water and reprecipitated with 150 ml. of cold acetone. After 10 minutes the precipitate was removed by centrifugation and washed with 150 ml. of an acidified mixture of acetone and ether (1 : 1 by volume containing 0.001 N HCI). This treatment converted the gelat- inous precipitate to a somewhat granular one which was then washed with 50 ml. of ether and then air-dried. The dried precipitate was dissolved in 35 ml. of cold water and the pH adjusted to 6.5. This solution contained 3.92 mmoles of total A5P as determined by the optical density, and 2.6 mmoles of adenyl acetate. The yield of adenyl acetate was 45 per cent, based on the amount of A5P used, but 66 per cent according to the A5P recovered. The supernatant fluid of the acetone precipitation contained the remainder of the A5P and some adenyl acetate, but no attempts were made to recover this material. The ratio of adenyl acetate to free A5P de- pends to a large extent on the rapidity with which the material is precipi- tated from the pyridine-containing mixture since pyridine catalyzes a rapid hydrolysis of adenyl acetate. Purifialion of Adenyl A cetate--Adenyl acetate was purified by anion exchange chromatography with Dowex 1 resin. In those preparations in which the ratio of adenyl acetate to A5P was 1 or less, a preliminary barium fractionation was employed. One typical fractionation experiment was carried out as follows: To 20 ml. of the crude adenyl acetate solution ob- tained in Procedure A were added 30 ml. of cold water, then 0.7 ml. of a saturated solution of barium chloride, followed by 175 ml. of 95 per cent ethanol. After 4 hours at -15') the precipitate of barium adenylate was removed by centrifugation and washed witch 60 per cent ethanol. The combined wash and supernatant fluids were adjusted to pE-1 6.5 to 7.0 and concentrated in vucw) to about 20 ml. to remove most of the ethanol. This solution was diluted to 150 ml. with cold water and put on a 2.5 X 5 cm. column of Dowex 1 C1- (200-400 mesh, 2 per cent cross-linked). The adenyl acetate was eluted almost immediately with 0.015 N HC1 (peak at 2 resin bed volumes). This procedure was followed when eluates suitable for enzymatic and chemical analysis were needed. By starting with 56 pmoles of adenyl acetate, 45.7 pmoles were eluted between 38 and 53 ml. of eluate. These three fractions contained 57.8 pmoles of A5P, indicating a purity of 79 per cent or a purification of almost 6-fold. The purity of the best prepa- rations obtained by the combined barium fractionation and chromato- graphic separation ranged between 75 and 85 per cent, based on the optical density at 260 mp and on the enzymatic activity. P. BERG 1019 Total A5P by optical density.. . . . . . . . . . . A5P deaminated by A5P deaminase.. . . . . Adenyl acetate by enzymatic assay.. . . . . Pi liberated by 5'-nucleotidaee. . . . . . . . . . . Results All of the preparations of adenyl acetate obtained as described above had an absorption spectrum indistinguishable from A5P. At pH 7 the absorp- tion maximum was at 259 mp and the X280/X260 and X250/X260 ratios were 0.18 and 0.86, respectively. In the samples of adenyl acetate prepared by Procedure B, only a small fraction of the A5P, determined from the ab- sorption spectrum, was present as free A5P. This was shown by the use of A5P deaminase which is specific for the 5'-phosphate ester of adenosine (5, 12) and by 5'-nucleotidase which is relatively specific for the 5'-phos- phate esters of nucleotides (6). Table I1 shows that only 18 per cent of the total A5P present was deaminated or dephosphorylated by these enzymes. Original pmles per ml. 4.5 3.2 0.86 0.84 TABLE I1 Enzymatic Analysis of Adenyl Acetate Prepared by Procedure B Hydrolyzed. pmoler per ml. 0.0 4.6 4.7 ~~ * The analyses were made on both untreated and hydrolyzed aliquots. Adenyl acetate was hydrolyzed by incubation with 0.01 N KOH for 5 minutes at room tem- perature. Total A5P was calculated from the measurement of optical density at 260 mp, and adenyl acetate was determined by enzymatic conversion to ATP. Free A5P was measured by phosphate liberation after treatment with 5'-nucleotidase under the conditions previously described (6) and by the decrease in optical density at 265 rng in the presence of A5P deaminase (7). Treatment of the adenyl acetate with 0.01 N KOH at room temperature for 5 minutes or with neutral hydroxylamine at 37" resulted in the destruction of its enzymatic activity and the liberation of free A5P. Of the A5P which is not susceptible to A5P deaminase and 5'-nucleotidase ("bound" A5P), approximately 70 per cent was accounted for as adenyl acetate by its en- zymatic activity. The remainder of t,he "bound" A5P, 0.6 pmole per ml., can best be accounted for as the diacetyl derivative of A5P. This can be seen in the following experiment (Table 111). 5 ml. of a solution contain- ing 100 pmoles of total A5P and 53 pmolcs of adenyl acetate per ml. were fractionated with Ba++ as described earlier and chromatographed on a Dowex 1 C1- column (2.5 X 5 cm.). The adenyl acetate was eluted with 0.0035 N HC1 at 4" and approximately 0.75 resin bed volume of eluate was collected per fraction. Ultraviolet-absorbing material started to appear in the eluate at about 6 resin bed volumes and continued to be eluted until 37 resin bed volumes had passed through the column, with the peak being at 1020 ADENYL ACETATE SYNTHESIS 25 resin bed volumes. Over this entire range the optical density ratio at 280 to 260 mp remained between 0.20 and 0.22. The fractions comprising the major portion of the peak (14 to 32 resin bed volumes) were analyzed as described in Table 111. It can be seen that there is good agreement be- tween the enzymatically determined adenyl acetate and the amount of acethydroxamic acid formed at 37", but each of these is lower than the total amount of A5P present, even though in the early fractions free A5P TABLE 111 Analysis of Chromatographed Adenyl Acetate ~~ Fraction No. 19 21 23 25 27 2!3 31 33 35 37 39 41 43 Free ASP 0.00 0.00 0.00 0.02 0.03 0.05 0.02 Adenyl. acetate 0.34 0.44 0.48 0.51 0.60 0.63 0.67 0.72 0.74 0.73 0.66 0.65 0.56 Acethydroxamatc Formed at 37' 0.33 0.40 0.47 0.50 0.57 0.62 0.66 0.70 0.73 0.70 0.66 0.63 0.53 Formed at 100' 0.56 0.75 0.93 1.01 1.10 1.02 0.82 Acetate 0.58 0.78 0.95 1.04 1.09 1-04 0.80 Total A5P Determined 0.43 0.54 0.63 0.71 0.79 0.84 0.91 0.97 0.99 0.96 0.92 0.84 0.70 Calculated 0.45 0.62 0.78 0.86 0.95 0.91 0.72 All of the values are expressed as micromoles per ml. Free A5P + adenyl acetate + (acethydroxamate, 100" - acethydroxamate, 37"/- 2) - total A5P. The fractions were analyzed for total A5P (by optical density at 260 mp), acethydroxamic acid formed at 37" and lOO", adenyl acetate (by conversion to acetyl CoA), free A5P (by A5P deaminase), and acetate (by acetokinase after in- cubation of the adenyl acetate in 0.01 N KOH for 5 minutes at room temperature). could not be detected. Moreover, there is more acetate liberated on treat- ment with dilute alkali than can be accounted for as adenyl acetate or acethydroxamate (37"). The data show, however, that all of the acetate present is converted to acethydroxamate at 100". If it is assumed that the extra acethydroxamate formed at 100' is derived from adenyl diace- tate, then all of the A5P can be accounted for. Thus, if one adds to the sum of free A5P and adenyI acetate one-half the difference between acethy- droxamic acid formed at 100" and 37", then there is good agreement with the value of total A5P. By Procedure B, it has been found that variable amounts of adenyl di- P. BERG 1021 acetate are formed and usually comprise 10 to 20 per cent of the total A5P of purified adenyl acetate preparations. The present evidence suggests that the diacetyl derivative is not enzymatically active in the formation of ATP or acetyl CoA, but more extensive studies with this compound sepa- rated from the monoacetyl derivative would be required to establish this point conclusively. Stabitity of Adenyl Acetale-Fig. 1 shows the kinetics of destruction of adenyl acetate under various conditions of temperature and pH. A sig- FIG. 1. Hydrolysis of adenyl acetate. Curve A, unbdered aqueous solution, 1W0; Curve B, 0.10 M sodium formate, pH 4.0,lOO"; Curve C, 0.10 M Tris buffer, pH 7.5, 100"; Curve D, 0.01 N HCI, 100"; Curve E, 0.10 N HCI, 100"; Curve F, 0.10 M glycine buffer, pH 10.0, 20", and 0.01 N KOH, 20". The firat order hydrolyeie con- stants calculated from Curves A, B, C, D, E, and F are 0.026,0.030,0.061,0.209,1.37, and 3.40 min.-l, respectively. nificant feature of this experiment was the comparative stability of adenyl acetate at pH 4.0 at loo", and the rapid destruction at pH 10 at 20". In the course of chromatographing adenyl acetate, little or no detectable disap- pearance of the adenyl acetate waa observed when it was kept at 0-4" in 0.01 N HC1 for as long as 12 hours, but even a minute's exposure at this temperature to a pH of about 10 or above results in the complete hydrolysis of the compound. When adenyl acetate was incubated at 37" in 0.10 M potassium phosphate buffer, pH 7.5, or in 0.10 M Tris buffer, pH 7.5, there was less than 10 per cent destruction in 30 minutes. However, in 0.10 M Tris buffer, at pH 8.0, 8.5, or 9.0, at 37", there was approximately 15, 30, and 50 per cent destruction in 30 minutes. 1022 ADENYL ACETATE SYNTHESIS DISCUSSION In the present report the preparation and some of the properties of adenyl acetate have been described. All of the evidence available thus far indi- cates that adenyl acetate is an anhydride of acetic and adenylic acids joined by an acyl-phosphate linkage. In support of this conclusion are the follow- ing observations. The absorption spectra of adenyl acetate preparations of up to 85 per cent purity are almost identical to that of free A5P. How- ever, the lack of reaction with A5P deaminase shows that the A5P moiety exists in a modified or "bound" form. Moreover, adenyl acetate is not degraded by 5'-nucleotidase, suggesting that the acetyl moiety is on the 5'-phosphate group. Treatment of adenyl acetate with hydroxylamine at neutral pH results in the liberation of A5P and the formation of acethy- droxamic acid. The acethydroxamic acid formed at neutral pH and room temperature is equivalent to the amount of acetyl groups which can be en- zymatically transferred to CoA. The stability of adenyl acetate under various conditions of pH and temperature closely resembles that of acetyl phosphate (13), the major difference being the greater stability at p1-I 4.0 at 100". All of these findings taken together are in agreement with the formulation of adenyl acetate as the phosphoacetyl derivative of A5P. The method of synthesis of adenyl acylates described by Avison (11), and used here for adenyl acetate, appears to offer a convenient method for the preparation of a number of similar derivatives. Peng2 has prepared adenyl butyrate from butyric anhydride and A5P and has shown it to be converted to ATP and butyryl CoA with the butyrate-activating system of liver (14). Comparable studies with the higher fatty acid adenylates are worthy of further investigation. SUMMARY Adenyl acetate has been synthesized by the reaction of acetic anhydride and A5P in pyridine and from acetyl chloride and silver adenylate. Studies with A5P deaminase and 5'-nucleotidase indicate that adenyl acetate is the phosphoacetyl derivative of A5P. In agreement with this conclusion as the finding of acethydroxamate and A5P formation on treatment of adenyl acetate with neutral hydroxylamine. Adenyl acetate is relatively stable at acid pH at 0-4", but is rapidly split at pH 2 and below, at 100". At pH 10 and above, it is rapidiy hydrolyzed even at 0". BIBLIOGRAPHY 1. Berg, P., J. Am. Chem. Soc., 77,3163 (1955). 2. Berg, P., J. Biol. Chem., 222,991 (1956). 3. Kornberg, A., J. Bid. Chem., 182,805 (1950). ~~ a Personal communication from Dr. H. Beinert and Dr. C. H. Lee Peng. P. BERG 1023 4. Ochoa, S., Stern, J. R., and Schneider, M. C., J. Biol. Chem., 195, 691 (1951). 6. Kalckar, H. M., J. BioZ. Chem., 167,461 (1947). 6. Heppel, L. A., and Hilmoe, R. J., J. Biol. Chem., 188,665 (1951). 7. Kalckar, H. M., J. Biol. Chem., 167, 445 (1947). 8. Fiske, C. H., and Subbarow, Y., J. Bid. Chenz., 68, 375 (1925). 9. Rose, I. A., Grunberg-Manago, M., Korey, S. R. , and Ochoa, S., J. Biol. Chem., 211, 737 (1954). 10. Lipmann, F., and Tuttle, L. C., J. Biot. Chem., 163,671 (1944). 11. Avison, A. W. D., J. Chem. SOC., 732 (1955). 12. Schmidt, G., 2. physio2. Chem. , 179, 243 (1928). 13. Koshland, D. E., Jr., J. Am. Chem. SOC., 74,2286 (1952). 14. MahIer, H. R., Wakil, S. J., and Bock, R. M., J. Biol. Chem., 204,453 (1953).