we find A S = AS.4
- 64 cal./dcg.
where ASA is the bracketed term above. Thus dEo/dT = AS/nF = A s A / n F - 1.39 mv./degree. The reference cell (S.C.E.) itself has a temperature coefficient of -0.56 mv./degree. Hence, with S.C.E. as the reference electrode
(15) nF [As* - l/*AS*] - 1.29 mv. The observed temperature coefficient of El,, is -1.03 mv. This leaves for the bracketed term a value of ca. 12 e.u., a small value considering the size of the molecules involved, and in keeping with the sinal1 difference in entropy between hydrogenated and olefinic hydrocarbons, and the small entropy of activation to be expected for the simple transition (probably involving a hydration) to stable DA. The coefficient a t pH 6.67 differs from that a t =
[CONTRIBUTION FROM
THE
pH 3.58 because of the more negative SH+term and also because the active hydrogen atom in the center of the molecule is dissociated a t this pH,2 so that A S must also include the entropy of dissociation. The first effect adds an additional -0.62 mv. to equation 15, the second a small positive increment, in view of the- small positive entropy of dissociation of most acids. The observed increment in dEl,,/dT 4s -0.4 mv. I t should be noted that the difference in temperature coefficient a t the two acidities cannot be accounted for by a mechanism in which an irreversible oxidation is the rate-determining step. To explain the observed relation between pH and El/, the hydrogen ions must be supposed to take part in a reversible equilibrium preceding the rate determining step. Thus the hydrogen ions are not involved in the formation of the activated cornplex, and cannot affect the activation energy, which would be the same a t both aciditics. CAMBRIDGE, MASS.
DEPARTMENT OF CHEMISTRY, IOWASTATE COLLEGE]
On the Mode of Interaction of Surface Active Cations with Ovalbumin and Bovine Plasma Albumin1 BY JOSEPH F. FOSTER AND JEN TSI YANG RECEIVED AUGUST6, 1953 The combinatioii of a cationic detergent, dodecyldimethylbenzylammonium chloride, with ovalbun~in(0)and bovine 1)I:tsma albumin (A) has been studied b y electrophoresis, equilibrium dialysis, viscosity mid optical rotation. 0 reacts it! an all-or-none rnanner accompanied by denaturation of the protein and a consequent increase in viscosity and optical rotat ion Determination of the number of ions bound is rendered difficult by the tendency toward gelation and precipitation of a part of the protein. Heat-denatured 0, and both A and heat-denatured A, fail t o show this all-or-none type of reaction, only a single electrophoretic coinpoaeilt bping observed. At the pH of the experiments (2.5-3) A is denatured, there being a nearly tenfold increase in specific viscosity (at low ionic strength) and a 30% increase in optical rotation even in the absence of the detergent. I n view of the absence of streaming birefringence and of the instantaneous character of the changes in viscosity and rotation it is coiicluded that this denaturation is essentially a n isotropic expansion of the molecule.
The combination of proteins with surface active anions has been the object of much study during the past decade. I t has been amply demonstrated that in the case of native ovalbumin (0)and horse and bovine serum albumin ( 4 ) an all-or-none type of combination occurs, a large number of ions being bound essentially as a nit.^-^ In the case of heat-denatured proteins, on the other hand, the binding is stepwise2q4suggesting that the all-ornone reaction is associated with a configurational change in the protein (denaturation). Studies with surface active cations are much more limited, in spite of the considerable biochemical interest in such compounds arising out of their use as bactericidal agents. Precipitation of proteins 011 the alkaline side of their isoelectric points by such compounds has been reported by several (1) Journal Paper Number J-1787 of the Iowa Agricultural Experiment Station, Ames, Iowa, Proj. 1223. Presented in part before the Division of Biological Chemistry of the American Chemical Society, March, 1953. This work was carried out under contract NonrSOS(O0) of the O 5 c e of Naval Research. (2) H. P. Lundgren, D . W. Blam and R. A. O’Connell, J . B i d . Chcm., 149, 183 (1943). (3) F. W. Putnam and H. Neurath, i b i d . , 169. 19.5 (1945). (4) J. T . Yang and J. F. Foster, THISJ O U R N A L , 76, 55GO (1953).
workers.6-8 In a more detailed study of the action of dodecylamine hydrochloride on 0, Timasheff and Nordg concluded that two stages of binding exist, corresponding to 30 and approximately 100 ions per mole of protein, respectively. In another study from this Laboratory it has been shown recently that several surface active cations denature 0 in acid solution. lo Studies by streaming birefringence and light scattering were in accord with the view that the reaction involves an unfoldingaof the 0 molecule to rodlets of length about 600 A. which then aggregate laterally. In the present study the nature of the interaction of alkyldimethylbenzylammonium chloride with both 0 and A is examined in more detail by electrophoresis, equilibrium dialysis, viscosity and optical rotation methods. (5) R. Kuhn, €1. J. Bielig and 0.Dann. Bar., 73B, 1080 (1940). (6) W. G. Jaffe, J . Bioi. Chcm., 148, 185 (1943). (7) K. H. Schmidt, 2. physiol. Chcm., ‘277, 117 (1943). (8) H. N. Glassman and D . M. Molnar, Arch. Biochcm. Biophys., 32, 170 (1951). (9) S. N. Timasheff and F. F. Nord. ibid.,31,309 (1951). (10) G. F. Hanna and I. F. Foster, J . P h y s . Chdm., 67, 614 (1053).
-D
A-
-D
A-
A I\
a -
I
I /
VA0.0
180 MIN,
I / A 0,I
166 MIN,
1/00.0 256 MIN. OI
I/O 0.1 205
MIN.
3
i
I / A 0 , 2 150 MIN, 01
A,'
y,
0
I 3
1/0 082 192 MIN, I/A0,3 Fig. l.-Typic;il
142 MIN,
elcctrophorctic pAttcrns of thc tlotii~cvldiniethylbenzylammoniumcomplexes of ovalbumin ant1 hovitic serum albumin. Electrophoresis was at 6.2 i 0 . 2 voltsjcni.
Materials and Methods Proteins.-Bovine plasma albumin ( A ) was obtained through the courtesy of Armour and Company. Ovalhumin (0)was prepared from fresh egg white by ammonium sulfate precipitation a t the isoelectric point, recrystallized three times and dialyzed free of salt. Buffer.-Glycine-Sac1 buffer was prepared with reagentgrade chemicals: glycine 0.09 -11,HCl 0.01 111 and S a C l 0.09 X,pH 3.3 and ionic strength 0.10. Cationic Detergent .--241ky1 (principally, x-dodecyl 1 diiriethylbenz!-lainmonium chloridr \vas purchased from Ouys Oil and Chemical Company. Electrophoresis .-Electrophoretic analyses were carried out a t 2.0' in the Tiselius-type cell, rvith a modified PhilpotSvensson cylindrical lens optical system. Equilibrium Dialysis.-Tnenty-ml. portions of buffered protein-detergent mixtures contained in Visking casings (?O ',2 inches in diameter) mere equilibrated against equal volumes of the buffer in glass-stoppered test-tubes, which ivere shaken gently at 1-3' for two days. The dialyzates \,.ere tllen diluted, if necessary, and analyzed spectrophototnetrically. Control evperinicnts with buffer only mere frequently carried out t o minimize the analytical errors. Spectrophotometric Analyses.-Both protein and detergent concentrations ivere determined with the model D U Beckman spectrophotometer. T h e extinction coefficients of A atld 0 E, m : , 6.70 at 279 m p and 7.50 at 280 mp, respectively. The molar extinction coefiicient of the detergent \vas found t o be 400 a t 263 mp." Viscosity.-The viscosity measurements: were made in an Ostwald-tvpe viscosimeter a t 24.9 L 0.1'. The flow time for water was 03 seconds. Optical Rotation.-The optical rotation measurements irere made a t 25 3" using a sodium lamp as light source.
~ ; 7 ~
+
Results and Discussion Electrophoretic Behavior.-In Fig. 1 are given examples of typical electrophoretic patterns ohtained with the detergent complexes of 0 and A. The ~al11eall-or-none behavior ~ h i c hcharacterizes ( 1 1 ) T T I-III,:, 2 n d
1
1:
l'
