6198
DAvm LIPKIN,~ ‘ I L L I A MH.
The difference in the amount of clottable protein in PBF and A B F can be due in part to the fact that the determinations on P B F were made after the sample had been stored in the lyophilized form in the refrigerator at about 5” for several years. The determinations on ABF were made on the fresh preparation since the preparation was never lyophilized. Clottability of fibrinogen does not indicate homogeneity either with regard to sedimentation or with regard to electrophoresis. I t would seem that one can obtain a homogeneous preparation of fibrinogen with low clottability. In the case of PBF, the original intent was to purify a quantity of fibrinogen to be stored iii the lyophilized form so that a number of studies could be made on the same material. Some clottability is lost during the drying process. Although the clottability of P B F was low, its physical properties are similar to those of ABF. By all standards ABF was a good preparation of fibrinogen. Furthermore, the sedimentation and electrophoretic homogeneities of both P B F and ABF were similarly improved on purification. The results for ABF agree rather well with those of Casassa3I and Sturtevant
ct nl.32 Both groups of workers found variations in the clottability of fibrinogen samples prci)arrd by purification of comnicrcial bovine fibriiiogcn. Summary Definite changes in the sedimentation, viscosity and light scattering properties of two preparations of bovine fibrinogen are observed in solutions of NaCl a t various ionic strengths. The changes are discussed in relation to shape, volume, hydration, and interaction of the protein molecules. The influence of the presence uf‘ impurities is also discussed. The molecular parameters were calculated from sedimentation, viscosity and partial specific volume data. The weight-average niolecular weights were caIcuIated from light scattering data. The average molecular weight from both sets of data a t an ionic strength of 0.1 is 249,000 for both preparations of fibrinogen. (32) J. hl. Sturtevant, A I Laskomski, J r , T I1 1)onnelly a n d H. A Scheraga, TIXIS J O U R K A I , 77, G 1 G 8 ( I O i i J
CHICAGO 11, ILLINOIS
(31) J. M. Casassa, J . Phys. Chem., 6 0 , 920 (1950).
-
[CONTRIBUTIOS FROM THE DEPARTVEST OF CIIEMISTRY, \T.’ASTIIXGTOS UXIVERSITY]
Adenosine-3’ :5’-phosphoric Acid : A Proof of Structure’ BY
DAVID
LIPKIN,WILLIAMH.
COOK’ AND R O Y L~ARI(H.\M’
RECEIVED MAY 0 , 1059 One of the products obtained in the degradation of adenosinc-~’-triphosphoricacid ( A T P ) in aqueous barium hydroxide a t 100” was shown conclusively t o be adcnosine-3’: 5’-phosphoric acid, a six-membered cyclic phosphate. This cornpound is identical with the cofactor for the interconversion of liver phosphorylase-dephosphophosphorylaseisolated from natural sources. It also has been found t o mediate t h e adrenocorticotropic hormone activation of adrenal cortical phosphorylase. T h e proof of structure of adenosine-3’:5’-phosphoric acid was obtained by t h e use of a variety of enzymatic and chemical degradations, as well as by t h e direct determination of t h e molecular weight of t h e compound by ultracentrifugation. Adenosine-3’: 5’-phosphoric acid was prepared also b y t h e action of dicyclohexylcarbodiimide on adenosine-j‘-phosphoric acid A new technique for degrading adenine nucleotides by means of liquid, anhydrous hydrogen fluoride is descrihed.
A new adenine ribonucleotide with unexpected criteria, they have demonstrated that this factor properties has been isolated from the products ob- and the A-3’:5’-P obtained by the degradation of tained on degradation of adenosine-5’-triphosphoric are identical.’ Haynes has found that Aacid (ATP) with aqueous barium h y d r ~ x i d e . ~ J3’:5’-P serves as an intermediate agent in the inThis new substance was assigned a cyclic dinucleo- duced stimulation of adrenal phosphorylase by tide structure as the result of a preliminary study. adrenocorticotropic hormone (ACTH).8 Further experiments reported in this paper show It was readily demonstrated that 12-3‘: 5’-P is conclusively that the original structural assignment an adenine nucleotide. The following observations is incorrect and that the new nucleotide actually is led to this conclusion: (1) I t s ultraviolet absorpthe monomeric cyclic phosphate, adenosine-3‘: 5’- tion spectrum is essentially identical with the specphosphoric acid (-4-3’:5’-P). tra of the adenylic acids: (?) deamination by Sutherland, et al., have reported the formation of nitrous acid yields a product with the characterisa heat-stable factor by tissue particles which stiniu- tics of an inosine derivative: (3’) adenine is oblates the interconversion of liver phosphorylase and tained on acid hydrolysis : (4) elementary analyses the dephosphophosphorylase.6 Using a variety of for carbon, hydrogen, nitrogen and phosphorus are those expected for an adenine nucleotide: and, ( 5 ) ( 1 ) Presented a t the 133rd Meeting of the American Chemical adenosine is slowly formed by the action of Cvofalus Society, San Francisco, Calif., April, 1958. (2) Cniversal hf atch Co. Fellow, 1058-1059. ailamanteus venom. A11 other evidence which was (3) On leave f r o m t h e Agricultural Research Council Virus Research obtained is in agreement with this conclusion. Unit, Cambridge, England, and who mould like to t h a n k t h e Wellcome Configuration of the Anomeric Carbon Atom.Foundation for a travel grant. The specific rotation of ;2-3’:5’-P5is the same in (4) W. 13. Cook, D. Lipkin and R. M a r k h a m , TIXIS J O U R N A L , 79, 3607 (1967). sign (negative) and order of magnitude as the spe(.5) D. Lipkin, R. M a r k h a m a n d W. H. Cook, ibid., 8 1 , GO75 ( l K 9 ) . (GI T. W. Rall, E. W. Sutherland and J. Berthet, .7. Bioi. C h ~ w . . 2 2 4 , 463 (1957); T. W. Rall and E. W, Sutherland, i h i d . , 232, IOC,3
(1958).
( i )(a) E. W.Sutherland and T. TV. Rall, TIIISJ O I J R S A I , , 79, XC03 (lY.57); ( b ) .7. Biol. C h e i z . , 2 3 2 , 1077 (19,jS). (8) I < . C. II:Lynes, Jr., ihid., 233, 1220 ( 1 9 5 8 ) .
Dec. 5, 1959
ADENOSINE-3’ : 5‘-PI-IOSPHORIC
cific rotation of other adenosinephosphoric acids,4 in which the configuration of the glycosidic linkage is beta. I n the case of diphosphopyridine nucleotide (DPN), configurational inversion of one of the anomeric carbon atoms changes the specific rotation from [ a I z 3 D -34.8’ (HzO, c 1) for the @- to [ a l Z 3+14.3’ ~ (HzO, c I) for the a-isomer. The specific rotation, [ a I z 3 D , of the nicotinamide mononucleotide from @-DPNis -38.3’ (HzO, c l), while that of the corresponding compound from a-DPN is $58.2’ (HzO, c 1).lo Furthermore, it is worth noting that the specific rotation of g-(@-D-ribofuranosy1)-adenine is negative ([a]”D -65.5’ (HzO, c 0.6),11 while that of the a-isomer is positive ( [ a ] D +24’ (HzO, c 0.63)).12 Additional evidence for the 0-configurat ion in A-3‘:5‘-P is that the compound also can b e prepared by the action of dicyclohexylcarbodi imide (DCC) on a trialkylammonium salt of adeno sine5’-phosphoric acid (A-5’-P) in anhydrous pyri dine solution. The A-3‘ :5’-P prepared in this way is not only identical by the usual criteria with the material prepared from ATP, but i t also exhibits the biological activity characteristic of A-3’ : 5’-P.13 It should be noted that the thymidine analog of A3’ : 5’-P (T-3’ :5’-P) has been prepared by the action of DCC on an anhydrous pyridine solution of thymidine-5’-phosphoric acid1*(T-5’-P). Degree and Positions of Esterification of the Phosphoric Acid.-It was readily demonstrated that A-3‘:5‘-P is a diester of orthophosphoric acid. The compound was not attacked by prostate phosphomonoesterase, but i t was slowly degraded either by whole snake venom or by a purified diesterase fraction from venom.15 The electrophoretic mobility on paper of A-3’:5’-P relative to A-5’-P a t various pH’s indicated that the compound has only one acid dissociation constant in the PH region 3.59.2. I n addition, direct potentiometric titration of A-3’:sI-P with barium hydroxide showed only one sharp break in the titration curve corresponding to apK‘,of 3.82. ThefactthatA-3’:5’-P hasahigh Rfin solvent A5 was a further indication that i t was not a monoester of orthophosphoric acid. Evidence concerning the positions to which the phosphoric acid is attached was obtained in a variety of ways. Deamination of A-3’:5’-P with nitrous acid yields inosine-3’ :5’-phosphoric acid (I3‘ : 5’-P). This same compound is obtained by digestion of inosine-5’-triphosphoricacid (ITP) with barium hydroxide. These observations clearly indicate that the phosphorus is not attached to the heterocyclic amino nitrogen. The resistance of A3’ : 5’-P to oxidation by periodate shows that in the conversion of A T P to A-3‘:5’-P the 2’- or 3’-hydroxyl of ATP has been substituted. As men(9) J. Baddiley in E. Chargaff and J. N. Davidson, “ T h e Nucleic Acids,” Academic Press, Inc., h’ew York, N. Y.,1565, Vol. I , p , 178. (10) N. 0. Kaplan, hi. hl. Ciotti, F. E. Stolzenbach and N. 12. Bachus, THISJOURNAL, 71, 815 (1055). (11) J. Davoll and B. A. Lomy, i b i d . , 7’3, 1650 (1961). (12) R. S. Wright, G. M. Tener and H. G. Khorana, Chemistvy 6’ I n d u s t r y , 554 (1957). (13) We wish t o thank Dr. Earl W. Sutherland for testing this sample for biological activity. (11) G. RZ. Tener, H. G Khoranu, I-the untreated venom. Degradation of Methylated A-3’: 5’-P to 2-0-Methylribose.--A X-ing. sample of A-3’: 5’-P was methylated exhaustively by tlie Purdie mcthod using absolute methanol as the rcnctiiiii rnediuin.16’iy T h e methylated ;2-3’:5’-P, which \vas recovered as a sirup, was transferred t o a polycthyletie test-tube and it wits dried in ‘i’ucuo over phosphorus pentoxide. \%‘hen liquid, anliydrous hydrogen fluoride was added t(J the dried sirup, it dissolved instantly. -4ftcr m m t ( J f the hydrogen fluoride was evaporated by a stream of dry air, thc tube and its contents were transferred t o a vacuum desiccator containing soda lime. -4fter remaining in the desiccator overnight, excess calcium carbonate and then water were added t o the residue in the polyethylene tube. Tlie mixture T V ~ Scentrifuged and t h e supernatant was rernoved for examination by paper chromatography and electrophoresis. The electrophoretic and chromatographic d a t a on the carlx~liydrntein the supernatant are summarized in Table I. T h e data clearly indicate t h a t 2-0-methyl-ribose IS obtained by the degradation of the cxhaustively methylated -\-3‘:5‘-P. TABLE I P R O P E R T I E S OF O-h1ETHYLKIHOSE FROM THE DEGRADATION OF EXHAUSTIVELY METHYLATED d-3’:5’-P Electrophoresis M R , ~
7--Chromatography--y Cumpiund
21, , solvent F
hIeth1-lribose from *4-3’:5’-P 2-Deoxy-~-ribose
0.50 .47
Ri,
0.33 .31
. 49lg . 3419 .55lY .3B1Y .57 .40 .58‘S .40’9 D-Ribose ,41 .17 ,3719 ,1819 Mobilities relative t o D-ribose. 2-0-hIethylribose 3-0-hlethvlribose 5-0-hlethylribose
MR,~
fiH 9.2 (borate)
1.79 1.64
0.44 .42 . 4619 ,4919 .9019 1.01 0.99!9 1.00
solvent E solvent G
.. .. 1.21 1.00
Determination of Configuration of the Pentose .-.Approximately 1 ml. of liquid, anh>-drous hydrogen fluoride wab added t o 5 mg. of A-3’:5’-P in a polyethylene test-tube. (29) R. Markham in K Paech and A I . V. Tracey, “Modern Methods of Plant Analysis,” Springer-Verlag, Berlin, 1965, Vol. IV, p. 288
ACID
6201
T h e sdutiorx was thcii treated as described in the previous paragraph. Paper chromatography (\l;hatman No. 1 paper) of the supernatant and a known sample of D-ribose using solvent J (development time, 30 h r . ) 5howed t h a t the only pentose present in the supernatant was ribose.30 T h e Rfvalues of the other three pentoses relative t o ribose are lyxose, 0.74; arabinose, 0.50; and xylose, 0.66. Elcctrophoresis in 0.05 M borate buffer (PH 9.2) confirmed the identity of t h e pentose in t h e supernatant a s ribose. T h e electrophoretic mobilities (17 volts/cm., 3 hr.) of thc other three pentoses relative t o ribose are lyxose, 0.92; arabinose, 1.17; and xylose, 1.26. As controls, A-5’-P and A-2‘(3’)-1’ also were degraded b y liquid, anhydrous hydrogen fluoride, T h e pentoscs recovered in both of these experiments alsu were shown to be ribose. Evidence for tlie ribose configuratioti of the pentose in :\3‘:5’-P was obtained in two other ways. First, the nucleoside isolated by paper chromatography from a snake veiiom degradation of A-3! : 5’-P was oxidized rapidly in aqueous solution b y pcriodate. Pentose nucleosides which do il(Jt have a cis-glycol configuration are not rapidly oxidized uiitier the same condition^.^^ Second, t h e nucleoside-3’-phospIiate from t h e snake venom digestion of A-3’: 5’-P described in the previous section was separated b y paper chromatography (solvent B). It was dissolved in 300X of dry formamidc; 250X of 1 1i.f D C C in anhydrous pyridine was added t o the resulting solution.32 After 22 hr. at room temperature, another 250X of the D C C solution was added. After tliree more days the reaction mixture was chromatographed on paper (solvent 1).T h e fastest-moving and principal band, which was shown b y electrophoresis at PH 9.2 (borate) t l J be identical with authentic A-2’:3’-P, was isolated. It was dissolved in disodium phosphate-sodium hydroxide buffer, pH 12.2. After several hours at 37”, it was found by electrophoresis at pH 9.2 (borate) t h a t the hydrolysis of the cyclic phosphate from the D C C reaction was essentially complete. T h e hydrolyzate was subjected t o electrophoresis in borate buffer. T h e band which had the siinie mobility as authentic .1-2’(3’)-P was removed from the electrophoresis paper and was chromatographed on fresh paper using solvent I). Two ultraviolet-absorbing spots were found tvliich were identical with authentic X-2’-P and A-3’-P samples. Degradation of Mixtures of ATP a n d ITP by Means of Barium Hydroxide.-Mixtures of ATP and ITP in t h c ratio of 10: 1 were treated with 0.4 X barium hydroxide at 100O. T h e 10: 1 ratio was used in order t o ensure t h a t , if a rcactiotl occurred between two molecules of nucleoside triphosphnte, there was a good chance of a mixed adenosine-inosine compound being formed. On chromatography the reaction rnixtures were not found t o contain such a compou~itl,but they did contain the substances identified as 1-3’:5’-P and :\-3‘: 5‘-P. Specific Volume of A-5’-P and A-3‘:5’-P.--4 o.%g&g. sample of X-5’-P33 (phosphorus, 8.235;) mu dissolvctl iri water and tlie solution was made u p to a volunic of 25.00 nil. X “flask” p y ~ n o m e t e r approxiniatclL,~~ 15 inl. in V(JlUmC, was filled with this solution a t 24’. T h e weight ( i f the pycnometer plus solution was 34.1258 g. The weight o f the pycnometer filled with distilled water at 24’ \vas $l.O444 g. The empty pycriometer weighed 18.495fi g. [.sing these data. and a value of 1.0027 nil./g. for the specific v ~ i l u l i i co f water at %4’, the specific volunie of :\-5’-P was found tlJ be 0.53 ml./g. 4 0.2088-g. sample of A-3’: 5‘-P.l.T5 Ha0 (phosphorus, 8.5970) in 25 ml. of aqueous solution was.uscd for tlctermining the specific volume using the same pycnonietcr and procedure as for the A-5‘-P; pycnometer plus solution a t 24’, 34.0926 g. The specific volume of 6-3’:5’-P is calculated t o be 0.60 nil./g.35 (30) Ribose itself on being subjected t o the action of liquid, anhydrous hydrogen fluoride is converted in part t o what appears t o be a series of polyriboses. These give a positive test with the aniline phthalate spray. ( 3 1 ) J. J. F o x , N. P u n g , J. Davoll and G. B. Brown, Tars J O U R N A L , 7 8 , 2117 (1950). (32) M. Smith, J. G. hIoffatt and H. G . Khorana, i b i d . , 8 0 , 6204 (1958). ( 3 3 ) Sigma Chemical Co., St. Louis, hIo. (34) N. Bauer in A. U’eissberger, ”Physical R I e t l ~ o d sof Organic Chemistry,” Interscience Publishers, Inc , X e w Yurk, ti.Y . .1919. Part I, p. 2G5. ( 3 5 ) T h e specific volumes required for the calculation of molecular weights are those of the anhydrous compounds. Because of this, the
T h e ~ I J O V C values for the specific volumes are in agree- digest, arid t h a t the nuclcotitlc t1ebigri:itetl .\-3’-P i r i tlie ment with values found by Tennant and I’ilbr;tndt36 for a solvent D chromatogram wis tiot A-2’-P, 1n:itericil i i i the appropriate band from a borate electrophoresis w:is clironumber of different nucleic acid samplcs. They do not agree, er, with the values given by Cavalieri for tlie apparent matographed on paper using solvent A . T h e single iiltraiiiolar volumes of :1-2‘-P, A-3’-P and tlie isomeric cytidylic violet-absorbing band on this paper was removed m i l it w i s shown t o be pure A-3’-P by comparison with authentic .1-2‘acids .37 Ii’hile Cavnlieri’s values are very reliable relative to one another, their reliability on an absolute scale is quesP a n d A-3’-P by means of paper chromatography with solt ionable. vent D a s t h e developing solvent. T h e fact t h a t tlic ljariuiii Ultracentrifugal Determination of Molecular Weight ,22----4 hydroxide hydrolysis of :I-3‘: 5’-P yields A-S’-P and .~13’-P, Spinco analytical ultracentrifuge with schlieren optics was b u t no A-2‘-P, also was demonstrated by ineans of aiiioil excliange chroitiatography used for thc xnolecu1:ir weight determinations. Tlie .1-3‘: 5’-P solution used for the specific volume deterillination (PH 4) was used for the molecular weight deterDiscussion ininations by the Archibald method and by equilibrium +cdinientation. For these determinations the sample was The outstanding difference in cheniical prop,ubjected t o 280,000 g for i 2 hr. During tlie early part of erties between A-3’:5’-P arid A-2’:3’-P is their tlie sedimentation, f ~ i u rpairs of rneasurements were made of the concentration of .1-3’:5’-P a t tlic top and bottom sur- stability toward hydrolysis. The A-3‘ :5’-P, iii faces of the solutim, in addition t o a measurement of the which the phosphate is part of a six-riieiiibered concentration d i s t r i h t i o n a t equilibrium. A n additional value for tlie niolecular weight vas calcu- ring, is far more stable toward both acid and alkaI:ited from measurements of the diffusion coeficietit ( D ~ o40, ~ line hydrolysis than A-2’:3’-P. This difference iri X 1 0 - 7 ) and sedimentation coeflicient (.&,,w 0.3 S.) of A-3’:hydrolytic stability between five- and six-inenibered *?’-P(approx. 0.5% solution) in 0.2 A4 sodium chloride. cyclic phosphates has been reported ~)reviouslyfor The measureinent of tlie diffusion coefficient mas made from the rate of spreading of tlie boundary obt:tined iii a synthetic a number of other compound^.^'^^^ The stability of A-3‘ :5’-P toward hydrolysis indicates that the lioundary cell. Hydrolysis in Acid and Alkaline Solutions.--.In aqueous phosphate probably is not part of a strained strucsolution of A-3’:5’-P (8 nig./ml.) mas made 1 N in hydroture and that the ribofuranose ring is riot planar. chloric acid. I t was kept at 92” and aliquots were removed a t suitable time intervals. The aliquots were spotted on This coiiclusion is in agreement with the suggestio11 of F ~ r b e r gbased , ~ ~ on X-ray diffractiop data, that filter pnper and the papers mere developed in solvent B. The only ultraviolet-absorbing hyrlrolysis product which was the 3’-carbon atom in adenosine is 0.5 A. out of the olJ,served during the course of the reaction was adenine. plane of the ribofuraiiose ring. The fact that The rate of the reaction w:is measured by determining the A-3‘:5’-P can be prepared by the actioii of barium amount of cyclic phosphate remaining after various inter hydroxide on ATP, although iiiaiiy coriipcting vals and also from the amount of adenine present at various times. T h e d a t a clearly indicated t h a t t h e hydrolysis was reactions are possible to give various other prodfirst order i:i -1-3‘: 5‘-P with a half-life of 55 min. T h e reac- u c t ~ also , ~ argues for a relatively uristrained sixtion was follrwed for approximately six half-lives. Sutherland antl Rail7" found the half-life of A-3’:5’-P in 1 Nliydro- membered cyclic phosphate structure in A-3’ :3’-P. It is worth noting, however, that under coin parable chloric acid to be approximately 28 inin. at 100’. A portion of tlie same A-3’: 5’-P solutitrii as was used conditions the half-life toward hydroxide ioiiabove mas made 1 N in sodium liytlrrixide. T h e reaction catalyzed hydrolysis of 1. :2-isopro],ylidene-D-x~.l[)mixture was treated as above, except that :I temperature of Adenine W:LS tlie principal ultraviolct-absorb- furanose-3:5-phosphoric acid (4 hr.)?‘is about sevc‘ii product observed. I t wits found, however, times the half-life of A-3’:5’-P (36 i i i i i i . ) . t h a t during the course of the hydrolysis a small amount of The hydrolysis of *1-3’:5’-Pwith 1 A7 sodiuiii aticnosine appeared and then disappeared as the heating was hydroxide a t OSo yields adenine as the principd continued. Furthermore, the total amount of material in the reaction mixture absorbing at about 260 n1.u progressively ultraviolet-absorbing product. During the course tlecrexred durinc tlic course of t h c heating neriod. Tlie of the hvdrolvsis, the onlv intermediate which ai>-
.
-Inother sample of the solution was subjected t o paper electrophoresis at PH 9.2 (borate). Tlie ultraviolet-absorbing hands corresponding to the adenylic acids were eluted from tlie paper separately and the absorbance of the separate eluates determined. X-3’(2’)-P and A-5’-P were found t o be present in a molar ratio of 5 : l . Iii order t o demonstrate conclusively t!i:tt A-3’-P was present in t h e barium hydroxide quantity of material in t h e pycnometer has t o he adjusted t o take into account water of crystallization. I n doing this it also was necessary tu assume t h a t .%-3‘:5‘-1’ had a composition corre5ponding to t h a t of a n adenylic acid minus o n e molecule of water. This assumption, which is borne o u t by the results, does not in fact necessitate a large correction. (3G) €I. G. Tennant arid C. F. Vilbrandt, THISJ O U R N A L , 65, 427 (1943). ( 3 7 ) L. F . Cavalieri, i b i d . , 74, 5804 (1‘332); 76, 5’208 (1953).
trast to this behavior, paiitotlicnic aci(1-2:4phosphoric acid42 and 1:2-isopr:,I,~liderie-r,-Yylofuranose-3 :3-phosphoric acid2Lhj.tlrdyze in alkaline solution with the rupture oi the pliosphate ( 3 8 ) A bihliograpliy on this subject ih Kivrn i n r c r !I: 5i.e n l s o H. S . Xfosher, J. Reinhart and H. C. Prosier, 26:d , 7 5 , Ik‘J!l ( I < J Z 3 ) . , 29 ( 1 9 i 9 ) ; ,l
