Oct. 5 , 1957
5323
COMMUNICATIONS TO THE EDITOR
of amino acid provide a rational basis for the interpretation of patterns which may be attributed t o proteins in solution. (5) National Research Council Fellow. (6) American Cancer Society Fellow.
CONTRIBUTION No. 2238 GATESAND CRELLINLABORATORIES OF CHEMISTRY CALIFORNIA INSTITUTE O F TECHNOLOGY OLEG JARDETZKY6 PASADENA 4, CALIFORNIA CHRISTINED . J A R D E T Z K Y ~ RECEIVED AUGUST19, 1957 T H E ROLE OF DISULFIDE BONDS I N ANTIBODY SPECIFICITY Sir :
TABLE I EFFECT OF DISULFIDEREDUCTION O N ANTIBODYBINDING^ Expt.
Protein
Treatment
I
Y/C
x
lo-'
0.74 14.4 Dialysis A Anti-Lac .62 7.0 Detergent A Anti-Lac .70 5.6 Anti-Lac Detergent B .22 1.5 Anti-Lac Reduction B .13 0.8 Anti-L-I, Reduction B B Anti-L-I, Detergent .06 .3 B Anti-L-I, ....... . 00 .o B RY *G Reduction .14 .9 M 0 One ml. of protein solution approximately 2 X was dialyzed against 1 ml. of 4 X 10-6 M dye solution. The -SH content of the reduced proteins was measured by the amperometric titration method of Benesch, et aZ.* Different anti-Lac preparations were used in experiments A and B.
The occurrence of disulfide bonds in y-globulins3p4and antibody proteins6 has led us t o suggest6that these linkages play an essential role in the evident that detergent treatment alone reduces the maintenance of the specific configuration of the specific binding somewhat, about 2-fold, and that combining region of the antibody molecule. We the capacity for non-specific binding is acquired by have obtained evidence supporting this hypothesis the reduced proteins. When these effects are from experiments in which the disulfide bonds of taken into account the results demonstrate that the purified anti-hapten rabbit antibodies have been specific binding is greatly reduced, to the extent of reduced and the resulting sulfhydryl groups pre- about 7-fold, when reduction of the disulfide bonds vented from re-oxidizing. The effect of such re- occurs. The residual specific binding observed duction on the specific combination of the antibody may be due t o the fact that only about 10 disulfide with an homologous azohapten provided the basis bridges, out of a minimum content of 20,5 were for our conclusions. split in our procedure. The protein was reduced with 0.1 M P-rnercaptoThe additional negative charge acquired by the ethylamine-HC1 a t $H 7.4 in the presence of reduced antibody by reaction of the -SH groups 0.1 Jf sodium decyl sulfate. The reducing agent with iodoacetate probably does not play a major role was removed by passage of the reaction mixture in the reduction of specific binding. Reaction of through a column of Dowex 50-X8(Na+). The reduced antibody with iodoacetamide shows the effluent reacted with an excess of iodoacetate by same results but such a preparation is insoluble a t overnight stirring a t room temperature. These p H 7.4 and is therefore less useful than the iodoaceoperations were done under anaerobic conditions tate derivative. by a procedure which will be described in detail R. E. Benesch, H. A. Lardy and R. Benesch, J . B i d . Ckem., 216, elsewhere. The protein solutions were subjected 663(8)(1955). to exhaustive dialysis against 0.001 M phosphate DEPARTMENT OF PEDIATRICS buffer p H 7.4 for the removal of the detergent. SCHOOL OF MEDICINE FREDKARUSH The low ionic strength was necessary to avoid the UNIVERSITYOF PENNSYLVANIA AND THE precipitation of the protein derivative a t this pH. CHILDRES'SHOSPITALOF PHILADELPHIA PA. The test antibody was that specific for the p- PHILADELPHIA, RECEIVED AUGUST21, 1957 azophenyl @-lactoside group (anti-Lac)' and the control proteins were rabbit y-pseudoglobulin (RypG) and antibody specific for the L-phenylDETERMINATION OF T H E SITE OF I4C I N (p-azobenzoylamino)-acetate group (anti-L-I,) .6 HYDROCORTISONE-14C DERIVED FROM The ability of the reduced proteins and various CHOLESTEROL-21-'4C INCUBATED WITH BOVINE ADRENAL GLAND TISSUE' control preparations to bind the azohapten 9-(pdimethy1aminobenzeneazo)-phenylP-lactoside (Lac Sir: dye) was measured by equilibrium dialysis a t 25' The biochemical conversion of cholesterol to the in 0.001 M phosphate buffer, p H 7.4. The results CP1-steroidsof the adrenal cortex was first demonare shown in Table I in terms of 7 and Y / C where r strated by investigators2 using cholester01-3-~~C. is the average number of dye molecules bound per Later other workers3 reported the isolation of a protein molecule a t the free dye concentration c. labeled six-carbon fragment resulting from the The last column, headed r / c , provides the most biochemical degradation of cholester01-26-~~Cby useful measure of the binding affinities. It is mammalian tissue extracts. The latter study suggested that the steroid hormones of the adrenal (1) These studies were aided by a grant from the National Science cortex could be derived from cholesterol involving Foundation and by a research grant (H-869) from the National Heart Institute of the National Institute of Health, Public Health Service. a degradation of the last six carbon atoms of the (2) I am indebted to Mrs. F. Karush for technical assistance in this (Cd. I. side chain [C,r.C,l iinvestigation, To obtain further experimental evidence on the (3) E. L. Smith and B. V. Jager, Ann. Rev. Microbiol., 6, 214 (1952). (4) 0. Markus and F. Karush, THISJOURNAL, 79, 134 (1957). (5) E. L. Smith, M. L. McFadden, A. Stockell and V. BuettnerJanusch, J . B i d . Chem., air, 197 (1955). (6) F. Karush, THISJOURNAL,18, 5519 (1966). (7) F. Karush, ;bid,, 79, 3380 (1967).
(1) This investigation was supported by the John J. Morton Cancer Fund and by a fellowship (HF-6137) from the National Heart Institute of the Public Health Service. (2) A. Zaffaroni, 0. Hechter and G. Pincus, THISJOURNAL, 73, 1390 (1951). (3) W.S,Lynn, Jr., E,Staple and S. Gurin, i b i d . , 76, 4048 (1964).
