AGNESEPP, T. RAMASARMA .4ND L. R. WETTER
721
affect an electrostatic repulsion dependent upon the charge density of the complexing anion. In order to conform to the observation that increasing the ionic strength of the same salt does not affect the energy of activation, we would have to postulate that these sites are fully saturated a t low salt concentrations. The following scheme is the summary of the proposed mechanism of collagen formation and its reversal T’
T
J_ T*
Collagen
where T is tropocollagen, T’, the inactive form and T*, the proposed intermediate that is being formed during the lag period. T’ is formed when active sites on tropocollagen are blocked by anions, cations or by such agents as urea. Conditions which enhance hydrogen bonding favor the formation of collagen, while those which rupture hydrogen bonds reverse the equilibrium. The equilibrium rate will be influenced by pH, type and ionic strength of the salt present, concentration of the
[CONTRIBUTION FROX THE NATIONAL
1701.
so
protein and temperature. Thus, when the collagen solution is removed from the presence of the fiber and conditions are adjusted to favor collagen formation, the system will remain poised unless conditions also favor a reasonable rate of reaction. I n writing this paper, we have avoided the use of such terms as neutral-soluble or acid-soluble collagen. The impressions gained from this work have cast considerable doubt in our minds as to whether such differences really exist. I t must be realized that when HOAc solutions of collagen are adjusted to values above pH 7, the optimum pH of fiber formation is being approached. If the other variables are such as to favor coagulation, fiber formation will occur. If not, a collagen solution which is soluble in neutral or alkaline solutions will be obtained. We have found that after coagulation, the fibers can be redissolved in HOAc, the collagen reneutralized, and fibril formation repeated as before. CLEVELAND 6, OHIO
RESEARCH COUNCIL
O F CANADA, PRAIRIE
REGIOXAL LABORATORY]
Infrared Studies on Complexes of Mg++with Adenosine Phosphates1 B Y AGNES EPP, T. RAMASARMA~ AND L. R.’IvETTER RECEIVED AUGUST5, 1957 Infrared spectra were obtained for the mono-, di- and triphosphate of adenosine and inosine in the presence and in the absence of I\lg++. The presence of the divalent cation affected the pyrophosphate group and the purine nucleus of the diand triphosphates of adenosine. The data suggest that these divalent ions complex with the abovc groups but no direct evidence is obtained as to whether the complex is intramolecular or intermolecular.
OH
Introduction Certain inorganic cations, notably Mg++, are essential for enzymic transphosphorylations involving the adenylic acid ~ y s t e m and , ~ a complex of Mg++ and ATP4 is the substrate for many kinase^.^-^ It has been suggesteds that complexes of this type involve the cation and the pyrophosphate group. Recently, however, Szent-Gyorgyis has postulated a quadridentate chelate between ATP and -1/Ig++, as in I, in which the bonding involves the 6-amino group, the nitrogen atom a t position 7 and one oxygen atom of each of the terminal phosphate groups. (1) Issued as N.R.C. KO,4576.
(2) Postdoctorate Fellow of t h e Xational Research Council of Canada, 1956-1957. Institute for Enzym- Research, University of Wisconsin, Madison. Wisconsin. (3) H. A. Lardy, “Phosphorus Metabolism,” Vol. I , ed. by U’.D. McElroy and B. Glass, T h e Johns Hopkins Press. Baltimore. l f d . , 1951, pp. 477. (4) T h e following abbreviations are used: ATP, adenusinetriphosphate; ADP, adenosinediphosphate; A M P , adenosinc-:’-monophosphate: I T P . inosinetriphosphate; I D P , inosinediphosphate and I M P , inosinemonophosphate. ( 5 ) H . G. Hers, Biochim. Biophya. Acta, 8 , 418, 424 (1952). (6) S. A. K u b y , L. Noda and H. A. Lardy, J . B i d . Chem., 210, 65 (19,54), (7) T. Ramasarma and L. R. Wetter, C a n . J . Biochen. Physiol., 36, 853 (1957). (8) C. S‘ruberg a n d I. Mandl. Avch. Biochcm., 23, 499 (1LJ49) (9) -4.Szent-Gyorgyi. “Enzymes: Units of Biological Structure and Function,” ed. b y 0. H. Gaebler, Academic Press, Inc., ISew York, S. Y., 1936, pp. 393.
I CH--C--C--C--C-O-P=O I
I
I
I
I
I
-0Such a chelate would be expected to give a characteristic infrared absorption spectrum distinct from that of uncomplexed ATP. Infrared spectra of nucleic acids have been recorded earlier and (10) E. R. Blout and M. Fields, Science, 107, 252 (1948). (11) E. R. Blout and M. Fields, J . Biol. Chem., 178, 335 (1949). (12) E. R . Blout a n d M. Fields, THISJ O U R N A L , 72, 479 (1950). (13) M. F. Morales and L. P. Cecchini, J . Celliiiov C o m p . Physiol., 37, 107 (1951). (14) H. Lenormant and E. R. Blout, C o n t p f .rend., 239, 1281 (1954). (15) E. R. Blout a n d H. Lenormant, Biochim. et Biophys. Acta, 17, 325 (19.52).
SPECTRA OF COMPLEXES OF MG++WITH ADENOSINE PHOSPHATES
Feb. .5, 1958
725
some general assignments related to their structural characteristics made. The present study examines the validity of the Szent-Gyorgyi formula I by noting the changes induced in the infrared spectra of adenosine phosphates by Mg++ and related cations. A related compound, inosine phosphate, also was investigated. Experimental ATP was purchased from Pabst Brewing Co., Milwaukee, as the crystalline disodium salt. ADP, AMP, I T P , I D P and I M P were obtained from Nutritional Biochemical Corp., Cleveland, and were used without further purification. The compounds were dissolved in water and the p H of their solutions was adjusted to 7.0 with XaOH. Magnesium ion was added where indicated as a n aqueous solution of MgC12. Samples for assay were freeze-dried with 5 ml. of 10% KBr solution and the powder pressed into a transparent window for infrared recordings.'e The spectra were measured with a Perkin-Elmer, Model 21, infrared spectrophotometer equipped with a sodium chloride prism. The major absorption bands of significance for the series of compounds investigated are given in Table I. The bands assigned to the phosphorus-containing groups are based primarily on the data of Corbridge.lg Other group assignments are from tables prepared by Bellamy20 and Jones and Sandorfy.2l
l
1
I
,000
l
I
l
I
PO00
I ,OD
WAVE
Fig. 1.-Infrared
1
NUMBER.
1
,
-
1-
-
1
.
'OD0
1
I00
(cm-0
spectra of (A) ATP alone, (B) equivalent amounts of ATP and Mg++.
C=N frequency. The nitrogen a t position 7 and the amino group a t 6 will be positively charged if they take part in complex formation with divalent cations, since electrons are being shared with the chelating metal ion. Thus the appearance of the band a t 1685 cm.-' probably is due to the C=N+ a t position 7 and suggests that the purine ring is Results involved in the chelation. Spectrum of ATP.-The infrared spectrum of Spectrum of ADP.-The spectrum of ADP reATP, with and without Mg++, shows three absembles that of ATP in all respects except that sorption bands in the region 900-1300 cm.-' (Fig. the POT band is absentz6 and the P=O band is 1): the pyrophosphate (P-0-P) group near 900 slightly weaker. The addition of Mg++ to ADP cm.-', the and C-0 (1050-1150 cm.-I), causes a shift of 15 cm.-' in P-0-P band toward the P=O group near 1250 cm.-' and a band of higher frequencies, and the C=N+ band appears medium intensity a t 980-990 cm.-l which is as- a t 1680 cm.-l (Table I). Spectrum of AMP.-The spectrum of AMP difsigned to the terminal POT group. Addition of Mg++, &In++or Co++ causes no marked changes fers considerably from those of ATP and ADP. in this region except that the P-0-P and the PO& The absorption bands caused by P-0-P and P=O bands are shifted by 10 cm.-' toward higher fre- groups are absent. However, the band a t 975 quencies (Table I). cm.-' assigned to the POT group is strong and well I n the 1500-1600 cm.-l region ATP exhibits a resolved. The absorption band a t 1075-1150 peak a t 1600 cm.-l and a shoulder a t 1575 cm.-l cm.-' caused by and C-0 as well as the N H which likely are due to the C=C and C=N stretch- deformation vibration band a t 1640 cm.-' are presing vibrations of the purine ring. Their intensities ent. The addition of Mg++ does not result in the are slightly decreased by the addition of Mg++ but appearance of the C=N+ band for AMP thus intheir positions are not altered. dicating that no complexing between the nitrogen A strong band appears a t 1640 cm.-' which is and the cation has taken place. However, there assigned to the N H deformation vibration of the is a shiit of the band a t 975 to 1000 cm.-' indicat6-amino group. The addition of divalent cations ing that some interaction takes place between to ATP results in the appearance in this region of a Mg++ and the phosphate group. second band a t 1685 cm.-' of medium intensity. Spectrum of AMP in Presence of Inorganic PyroAccording t o Lenormant and B l o ~ t ~this * , ~band ~ appearance of the C = k band can be caused by rearrangements of double bonds phosphate.-The within the purine nucleus. Other workers22-2j (1685 cm.-I) in the ATP and ADP spectra in the have reported that a positive charge on the nitrogen presence of M g + + and its absence in the AMP spectrum suggests that the P-0-P group plays a atom of the C=k group gives rise t o a higher role in the chelation of divalent cations. The spec(16) R. I,. Sinsheimer, R. L. N u t t e r and G. R. Hopkins, Biochim. tra of AMP with an equimolar amount of inorganic el R i o p h y s . Acta.. 18, 13 (1955). pyrophosphate and of adenosine with an equimolar (17) H. T. Miles, ibid., 22, 247 (1956). amount of inorganic pyrophosphate or tripoly(18) U. Schiedt and H. Reinwein, Z. Naltwforsch, 7b, 270 (1952). (19) D. E. C. Corbridge, J. A p p l . C h ~ m .6, , 456 (1956). phosphate are equivalent to the sum of the in(20) L. J. Bellamy, "The Inirared Spectra of Complex Molecules," dividual compounds. The addition of &Ig++ t o hlethuen and Co., Ltd., London, John Wiley a n d Sons, Inc., New these combinations does not result in the appearYork, N . Y., 1954. (21) N. R. Jones and C. Sandorfy, "Technique of Organic Chemistry, ance of the C=N+ band which suggests that chelaVol. IX, Chemical Applications of Spectroscopy," ed. by W. West, tion will not take place unless adenosine and P-0-P Interscience Publishers, Inc., New York, N . Y., 1956, pp. 247. are in the same molecule. (22) J. D. S. Goulden, J. Chem. Soc., 997 (1953).
6-6
5-6
(23) 0.E. Edwards, F. H. Clarke a n d E. Douglas, Can. J . Chem., 32, 235 (1954). (24) B. Witkop a n d J. B. Patrick, THISJOURNAL, 76, 4474 (1953). (2:) N. J Leonard and V. W. Gash, ibid., 76, 2781 (1954).
(26) T h e PO; band was present in the spectra of ArP, A M Y , ITP a n d I M P biit not A D P or I D P . It was noted t h a t the spectrum for sodium pyrophosphate showed a very weak PO; band whereas the spectrum for sodium tripolyphosphate gave a sharp b a n d a t 9 7 0 c m.-I.
7%
.!LGN~S EPP,
MAJORISFRARED .~BSORPTIOU BAXDSOF 1
?’. l
