Biochimica et Biophysica Acta, 578 (1979) 68--75 © Elsevier/North-Holland Biomedical Press
SODIUM DODECYL SULFATE-PROTEIN POLYPEPTIDE COMPLEXES IN 8 M U R E A WITH SPECIAL R E F E R E N C E TO SODIUM DODECYL SULFATE-POLYACRYLAMIDE GEL E L E C T R O P H O R E S I S
TOSHIO TAKAGI and KANENOBU KUBO
Institute for Protein Research, Osaka University, Suita, Osaka 565 and Faculty of Pharmacy, Kinki University, Higashiosaka, Osaka 577 (Japan) (Received October 4th, 1978)
Key words: SDS-protein complex; Protein denaturation; Urea; (Gel electrophoresis)
Summary The effects of 8 M urea on the complexes formed between sodium dodecyl sulfate and protein polypeptide were found to be as follows: (1) The maximum a m o u n t of SDS bound is reduced by almost half, and the minimum equilibrium concentration of SDS necessary to reach saturation was nearly doubled; (2) The apparent content of a-helical structure deduced from CD measurement is reduced to only 50--70% of that in the presence of sodium dodecyl sulfate alone; (3) The effective size of the sodium dodecyl sulfate-protein polypeptide complex deduced from viscosity measurements is increased, b u t is still smaller than the effective size of the protein in 8 M urea alone.
Introduction Both sodium dodecyl sulfate (SDS) and urea have been extensively used as strong protein denaturants. Polyacrylamide gel electrophoresis of proteins is often carried o u t in the presence of both of the t w o denaturants. Swank and Munkres  recommended the addition of 8 M urea to the buffers used in SDS-polyacrylamide gel electrophoresis, since the presence of both 8 M urea and SDS both improves the resolution of the system for proteins of molecular weight less than 10 000 and facilitates dissociation of protein aggregates [ 1--4]. To evaluate data obtained from SDS-polyacrylamide gel electrophoresis in A b b r e v i a t i o n s : P r o t e i n p o l y p e p t i d e , single p o l y p e p t i d e derived from a p r o t e i n and d e v o i d and native c o n f o r m a t i o n as t h e result o f binding t o SDS, with disulfide cross-linkages r e d u c e d and c a r b o x y a r n i d o m e t h y l a t e d , if t h e y w e r e p r e s e n t in t h e p r o t e i n ; R C A M , r e d u c e d and c a r b o x y a m i d o m e t h y l a t e d ; SDS, s o d i u m d o d e c y l sulfate.
69 the presence of 8 M urea, it is necessary to understand the effect of urea on the complexes formed between SDS and a protein polypeptide. The complexes formed in the absence of urea have been extensively studied [ 5--11 ]. No study has been made, however, on the complexes in the presence of urea. This article describes the results of comparative experiments on the states o f protein polypeptides in aqueous solutions containing either SDS or urea or both. Experimental Sodium dodecyl sulfate (SDS) designated as PSP-4 was obtained from Nakarai Chemicals as a specially prepared reagent. It was 97% pure by gas chromatography, the major contaminants being the decyl and tetradecyl sulfates. D o d e c y l t r i m e t h y l a m m o n i u m chloride (special grade) was obtained from T o k y o Kasei Chemicals. Urea (special grade) was obtained from Wako Chemicals, and used w i t h o u t further purification. Other reagents including proteins were special grade or those routinely used in polyacrylamide gel electrophoresis. Proteins with disulfide groups were reduced and carboxyamidomethylated (RCAM) according to the m e t h o d described previously . Protein concentration was determined spectrophotometrically using authentic values of Eleml~ at 280 nm (6.7, 7.4, 9.1, and 26.9 for bovine serum albumin, ovalbumin, ~-lactoglobulin, and lysozyme, respectively) or at 410 nm (90.7 for c y t o c h r o m e c). The concentration of hemoglobin was determined by the m e t h o d of Van Kampen-Zijlstra  as ferrihemoglobin cyanide, assuming the molar extinction coefficient for the heme group to be 7290 at 503 nm; the absorption was not affected by the presence of SDS at this wavelength. Sodium phosphate buffer of pH 7.0 was prepared by mixing 0.05 M, NaH2PO4 and 0.05 M Na2HPO4 and adding sodium azide to a final concentration of 0.02%; hereafter, this buffer will be referred to as 0.05 M sodium phosphate buffer, pH 7.0. When urea was added to this buffer to a final concentration o f 8 M, the pH value increased to 7.6. The critical micelle concentration of SDS in the presence or absence of 8 M urea was determined b y a m e t h o d utilizing the marked change of electrophoretic mobility of SDS on micelle formation in a polyacrylamide gel. SDS was added only to the upper reservoir of the electrophoretic apparatus and several runs were carried o u t at various SDS concentrations around that expected to be the critical micelle concentration. The relative mobility of SDS was estimated b y measurement of the position of a white band formed when the zonal front of the dodecyl sulfate ion ran counter to that of the dodecyltrimethylammonium ion which had been added to the lower reservoir. The concentration of the cationic surfactant was kept constant just below its critical miceUe concentrations throughout the runs. The binding of SDS to a protein polypeptide was measured b y the equilibrium dialysis technique, essentially according to the method described previously . Dialysis was continued for 14 days at 25.0 +'0.2°C. When the final equilibrium concentration of SDS was above the critical micelle concentration, the starting condition of equilibrium dialysis was brought close to the expected final condition. SDS was analyzed by colorimetry of the Methylene Blue-SDS complex extracted with chloroform .
70 CD spectra were measured with a JASCO J-20 CD spectrophotometer. The temperature of the sample solution was controlled at 25 _+0.1°C using a brass muffle obtained from JASCO. CD data are reported as mean residue ellipticity. Mean residue weight was assumed to be 112 for all proteins used. Sample solutions containing SDS were prepared in the same manner as in viscosity measurements. Viscosity was measured with a Ubbelohde type viscometer with flow time for water of 200 s. Protein samples were dissolved to a final concentration of between 2 and 4 mg/ml in 0.05 M sodium phosphate buffer, pH 7.0 containing 3.5 mM SDS and/or 8 M urea. Enough SDS was further added to the solutions to supply sufficient SDS for binding to the proteins; the required amounts of SDS were calculated from corresponding binding isotherms. A sample solution thus prepared was dialysed against a hundred volumes of an appropriate buffer. Dialysis was carried out for 14 days for the buffer containing SDS and several days for t h a t lacking it. For dilution of a sample solution in the viscometer, the outer solution was used as the solvent. Both the inner and the outer solutions were filtered through a Millipore Filter before pouring into the viscometer. Cytochrome c and hemoglobin were dissolved in the same manner as described above, and heated at 100°C in boiling water for 3 min before starting dialysis; other proteins were reduced and carboxyamidomethylated to make the above denaturation procedure unnecessary. Results
Protein polypeptides Sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis is usually carried out with proteins in which the intra- and inter-chain disulfide groups have been reductively cleaved. In t h e present study, sample proteins having disulfide groups were reduced and carboxyamidomethylated to prevent reformation of disulfide groups during preparation and measurements. Cytochrome c and hemoglobin, which lack disulfide groups, were simply heated at 100°C in the presence of necessary amounts of SDS and/or 8 M urea to complete denaturation.
Binding isotherms Fig. 1. shows binding isotherms of SDS to RCAM-lysozyme and RCAMbovine serum albumin in the presence and the absence of 8 M urea. The presence of 8 M urea had the following major effects on the binding isotherms: (1) The m a x i m u m a m o u n t of SDS bound was reduced by almost half; (2) The m a x i m u m equilibrium concentration of SDS necessary to attain maximum binding was nearly doubled. These effects have also been observed with two kinds of plant virus coat proteins by Sano et al. . The above effects of urea, therefore, may be taken to be c o m m o n to binding isotherms o f SDS to various kinds of protein polypeptides. RCAM-lysozyme was precipitated when the equilibrium concentration was below half of the critical micell concentration either in the presence (curve 2) or in the absence (curve 1) of 8 M urea. Presumably the complex formed between SDS and the highly basic RCAMlysozyme at low binding ratio is highly aggregated, making it insoluble even in 8 M urea.
1 p -
ppt~--I D ~0
jill! I "
,, y i
,i, 1 2 3 CONCENTRATION
4, j 4 OF $ D S ( m M )
Fig. 1. B i n d i n g i s o t h e i m s o f S D S t o p r o t e i n p o l y p e p t i d e s in t h e p r e s e n c e ( c u r v e s 2 a n d 4) a n d t h e a b s e n c e ( c u r v e s 1 a n d 3) o f 8 M u r e a in 0 . 0 5 M s o d i u m p h o s p h a t e b u f f e r , p H 7 . 0 a t 2 5 . 0 ° C . a, R C A M - l y s o z y m e ; b, RCAM-bovine serum albumin. Arrows indicate the critical micelle concentrations of SDS in the p r e s e n c e ( r i g h t ) a n d t h e a b s e n c e (left) o f 8 M u r e a . C u r v e 3 is c i t e d f r o m ref. 7. D u e t o t h e f o r m a t i o n o f i n s o l u b l e m a t e r i a l in t h e r e g i o n i n d i c a t e d as ' p p t ' i n Fig. l a , m e a s u r e m e n t a t l o w e r c o n c e n t r a t i o n s w a s n o t possible. Fig. 2. C D s p e c t r a o f p r o t e i n p o l y p e p t i d e s i n 3 . 5 m M S D S ( c u r v e 1), in 3 . 5 m M S D S p l u s 8 M u r e a ( c u r v e 2), a n d in 8 M u r e a ( c u r v e 3) a t 2 5 . 0 ° C . a, R C A M - b o v i n e s e r u m s e r u m a l b u m i n ; b, R C A M - l y s o z y m e ; c, R C A M - o v a l b u m i n . T h e s a m e b u f f e r s o l u t i o n as in Fig. 1 w a s u s e d .
As has been described in a previous paper , the binding of SDS to a protein polypeptide proceeds in two steps, as shown by curve 3 in Fig. 1. As shown b y c u r v e 4 in Fig. 1, in the presence of 8 M urea the first step disappeared and the a m o u n t of b o u n d SDS increased m o n o t o n o u s l y over a wide concentration range of SDS. The same effect has been observed with the virus coat proteins . The present study was initiated with special reference to SDS-polyacrylamide gel electrophoresis, which is carried o u t in the presence of SDS in a concentration above its critical micell concentration. The binding process before attainment of the saturation value was, therefore, not further investigated in the present study. CD spectra Fig. 2 shows CD spectra of RCAM-bovine serum albumin, RCAM-lysozyme, and RCAM-ovalbumin in the presence of 3.5 mM SDS and/or 8 M urea. In 3.5 mM SDS (curves labeled 1), CD spectra for the three protein polypeptides showed patterns anticipated for a mixture o f a-helix and random coil with the a m o u n t of s-helix present ca. 50%, according to Greenfield and Fasman . In 8 M urea (curves labeled 3), CD spectra for the three polypeptides showed patterns anticipated for a random coil . In the presence of both 3.5 mM SDS and 8 M urea, CD spectra showed patterns intermediate between the above two cases. The presence of 8 M urea made measurements below 205 nm impossible. It is to be n o t e d that the CD patterns observed for the three polypeptides derived from different proteins are similar in each of the three different conditions. Such a convergence to a c o m m o n CD pattern has been reported for many SDS-protein polypeptide complexes by Mattice et al. .
A O A - - A -v w v
4,4 4,8 log MW
Fig. 3. Plots o f r e d u c e d viscosity v e r s u s c o n c e n t r a t i o n for R C A M - l y s o z y m e (a) a n d R C A M - b o v i n e s e r u m a l b u m i n ( b ) a t 2 5 . 0 0 ° C in t h e p r e s e n c e o f 3.5 m M SDS (o o), 3.5 m M SDS plus 8 M u r e a (X X), a n d 8 M u r e a (¢ --), r e s p e c t i v e l y . T h e s a m e b u f f e r as in Fig. 1 w a s used. Fig. 4. L o g a r i t h m i c p l o t s o f intrinsic v i s c o s i t y , ['r/], v e r s u s m o l e c u l a r w e i g h t f o r v a r i o u s p r o t e i n p o l y p e p t i d e s in t h e p r e s e n c e of 3 . 5 m M SDS (o o), 3 . 5 m M SDS plus 8 M u r e a (X ×), a n d 8 M u r e a (e e ) , r e s p e c t i v e l y . P r o t e i n s are, f r o m left to right, b o v i n e h e a r t c y t o c h r o m e c, b o v i n e p a n c r e a t i c r i b o n u c l e a s e A, h e n egg w h i t e l y s o z y m e , h u m a n h e m o g l o b i n , b o v i n e fl-lactoglobulin, r a b b i t m u s c l e lactic dehydrogenase, h e n o v a l b u m i n , Bacillus s u b t i l u s s - a m y l a s e , a n d b o v i n e s e r u m a l b u m i n .
Viscosity Fig. 3 shows plots of reduced viscosity of SDS-protein polypeptide complex versus concentration for RCAM-lysozyme and RCAM-bovine serum albumin. Such viscosity measurements were carried o u t with six other kinds of protein polypeptides. The reduced viscosity was almost independent of the concentration of each protein polypeptide. Intrinsic viscosities obtained are plotted in Fig. 4 versus molecular weights of the protein polypeptides. Discussion
The proposal by Swank and Munkres  of the addition of 8 M urea to the medium of SDS-polyacrylamide gel electrophoresis was followed b y the widespread use of this modified technique. The improvement gained by the modification may be ascribed to two major factors: the first is the change of the nature of the aqueous SDS solution by the addition of 8 M urea. The second is that the presence of 8 M urea makes it possible to prepare a transparent and highly cross-linked polyacrylamide gel with an improved performance for small-sized protein polypeptides. The present study concerns the first factor. SDS-polyacrylamide gel electrophoresis in the presence of 8 M urea has been carried o u t in various buffer solutions. In the present study, 0.05 M sodium phosphate buffer, pH 7.0, was the sole buffer solution used. We believe that similar conclusions will be obtained with other buffers as far as the comparison with behaviors of a protein polypeptide in the three denaturing media (3.5 mM SDS, 3.5 mM SDS plus 8 M urea, and 8 M urea) is concerned.
SDS binding The major premises of SDS-polyacrylamide gel electrophoresis concerning SD8 binding are as follows [5,7] : (1) SDS binding is saturated; (2) The intrinsic charge of a protein polypeptide is smeared out to give a constant surface charge density, as the result of the high and nonspecific affinity of SDS, to give a nearly constant binding ratio of near 1.5 (weight to weight basis). SDS-polyacrylamide gel electrophoresis in the presence of 8 M urea is generally carried o u t in an SDS concentration of 3.5 mM (0.1%) or above. Though a significant lateral shift of the binding isotherm is observed for RCAM-lysozyme and RCAM-bovine serum albumin (Fig. 1), the SDS binding is saturated in this concentration range. The first premise is still satisfied even in the presence of 8 M urea. Reduction of the maximum amount of SDS b o u n d almost by half was observed with both RCAM-lysozyme and RCAM-bovine serum albumin in the present study, and with t w o virus coat proteins by Sano et al. , and clearly jeopardizes the second premise. Tung and Knight [ 15] have demonstrated using a model system (maleylated and unmaleylated virus coat proteins) that SDSpolyacrylamide gel electrophoresis gives incorrect molecular weight for proteins with relatively high net charge. The presence of 8 M urea evidently promotes such a tendency to give an incorrect estimation, due to insufficient smearing o u t of intrinsic charge.
Conformation To understand the effect of urea on the conformation of an SDS-protein polypeptide complex, it is prerequisite to have acknowledge a b o u t its conformation in the absence of 8 M urea. Several models have been proposed for this kind of complex [5--11]. The popular 'rod-like model' of Reynolds and Tanford  was introduced to interpret the deviation of intrinsic viscosities of the complexes from those expected for c o m p a c t globules assuming them to be rigid ellipsoids from the beginning, and, therefore, is n o t realistic. Models proposed thereafter [6--9] laid emphasis on the felxible nature of the complexes. In the present paper, we will discuss the results obtained in the light of the 'necklace model' [6,7] proposed by the group including one of the present authors (T.T.). In summary, the necklace model is characterized as follows: (1) The polypeptide chain of a complex is essentially flexible; (2) dodecyl sulfate ions bind to a protein polypeptide to form a micelle-like cluster; (3) the polypeptide chain locally assumes a-helical structure. According to the results o f CD measurements by Mattice et al. , a-helical content rages from 30 to 50% for many kinds of protein polypeptides. CD spectra in Fig. 2 indicate that most (50--70%) of the a-helical structure present in 3.5 mM SDS still persists in solution containing both 8 M urea and 3.5 mM SDS. This was judged from the CD intensity at 220 nm which is generally regarded as a measure o f a-helical content, assuming that no a-helical structure was present in 8 M urea . It is surprising that the a-helical structure still persists to such an extent in 8 M urea which forces most protein polypeptides to expand to a randomly coiled state . Presumably the hydrocarbon moieties of SDS molecules associate around a protein polypeptide to afford an environment sequestered from the concentrated urea solution and
74 favor formation of hydrogen bonds to maintain the a-helical structure. Such an effect is expected only if SDS is b o u n d to a protein polypeptide in groups to afford an extended hydrophobic environment which allows formation of a-helical structure. This is precisely the situation assumed in the necklace model . The fact that urea and SDS act in opposite way on the b0 value of ovalbumin even in each other's presence was pointed o u t by Meyer and Kauzmann . The ORD parameter, b0, suggested the same situation concerning the a-helical content for ovalbumin as that suggested from the CD data of Fig. 2 for the three protein polypeptides. At that time, urea as well as SDS was assumed to rupture hydrophobic interactions . The interpretation of the observed phenomena was, therefore, quite confusing . Similar experimental results were subsequently obtained by Jirgensons . The intrinsic viscosity, a measure of of hydrodynamically effective size of a complex, decreased with the change of medium in the following order: 8 M urea, 3.5 mM SDS plus 8 M urea, and 3.5 mM SDS (Fig. 4). This suggests that the increase in the effective size of a protein polypeptide by the binding of SDS is superseded by the decrease due to formation of a-helical structure. The a-helical structure is presumed to be local, short-range, and intermittent to allow flexibility of the complex. A similar decrease of intrinsic viscosity has been observed when amylose in an alkaline aqueous solution assumes a helical structure on the formation of a complex with SDS [19,20]. All three logarithmic plots of intrinsic viscosity versus molecular weight of the polypeptide moiety of the complexes (Fig. 4) are almost parallel, having inclination around 0.7. The value indicates that the polypeptide chains of the complexes behave as flexible coils from macroscopic point of view  in each of the three denaturing media. Intrinsic viscosities of the complexes in the absence of urea have been measured, and plots equivalent to the b o t t o m one in Fig. 4 have been reported by Reynolds and Tanford  and Hamauzu et al. . The two reported plots and ours, however, differ from each other. The inclination obtained by the other t w o groups [5,22] were in agreement, and were around 1.2, apparently indicating that the complexes are asymmetric in support of the rod-like model of Reynolds and Tanford . We can offer no explanation for the difference between the previously reported results and ours, except to point o u t that the previous measurements were made in media of salt concentrations significantly lower than that of the buffer solution used in the present study. Our results (Takagi, T., unpublished) indicate that the viscosity of an SDS-protein polypeptide complex is quite sensitive to salt concentration. Detailed data and discussion o f the viscosity of such complexes will be described in a subsequent paper. It is to be noted in Fig. 4 that the intrinsic viscosity o f the complexes in 3.5 mM SDS alone levels off as the molecular weight becomes less than 15 000. This phenomenon must be closely related to the lowering of the resolution of SDS-polyacrylamide gel electrophoresis for proteins in the regions of molecular weight less than 15 000 . In the presence o f 8 M urea, no leveling off was observed, and the value of intrinsic viscosity was still variable in this region. This might be a major factor improving the resolution of SDS-polyacrylamide gel electrophoresis in the presence of 8 M urea. The plot was, however, less
monotonous and points were scattered more than either of other two plots in Fig. 4. This will lead to erroneous estimates o f molecular weight. We wanted to investigate the correlation between the effective sizes of complexes in 8 M urea and their molecular weights for small-sized protein polypeptides in more detail. Preparation of a sample solution of SDS-protein polypeptide complex for a protein with molecular weight, less than 10 000 is quite difficult due to the lack of appropriate dialysis membrane. We, therefore, used gel chromatography to investigate the problem, but found that all conventional gels for molecular sieving deteriorated in aqueous solutions containing both of 3.5 mM SDS and 8 M urea, making gel chromatography impracticable. The solution of this problem requires further work. Acknowledgements This study was supported in part by research grants from the Ministry of Education, Science, and Culture of Japan, to T.T. (Grant Nos. 321930 and 338044). The authors express their heartfelt thanks to Prof. T. Isemura of Kinki University for his encouragement throughout this study. They also thank Miss. K. Uemura, Miss. T. Fujii, and Mr.'Y. Kikuchi for their skilful assistance in experiments. References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Swank, R.T. and Munk.res, K.D. (1971) Anal. Biochem. 39, 462--477 Bacl~ach, H.L. and Hess, W.R. (1973) Biochem. Biophys. Res. C o m m u n . 55, 141--149 Abraham, G. and Cooper, P.D. (1976) Anal. Biochem. 73, 439--446 Downer, N.W., Robinson, N.C., and Capaldi, R.A. (1976) Biochemistry 15, 2930--2936 Reynolds, J.A. and Tanford, C. (1970) J. Biol. Chem. 245, 5161--5165 Shirahama, K., Tsujii,K. and Takagi, T. (1974) J. Biochem. 75, 309--319 Takagi, T., Tsujii,K. and Shixahama, K. (1975) J. Biochem. 77, 939--947 Wright, A.K., Thompson, M.R. and Miller, R.L. (1975) Biochemistry 14, 3224--3228 Mattice, W.L., Riser, J.M. and Clark, D.S. (1976) Biochemistry 15, 4264--4272 Rowe, E.S. and Steinhardt, J. (1976) Biochemistry 15, 2579--2585 Dunker, A.K. and Kenyon, A.J. (1976) Biochem. J. 153, 191--197 V a n Kampen, E.J. and Zijlstra,W.G. (1961) Clin. Chim. Acta 6, 538--544 Sano, Y., Nozu, Y. and Inoue, H. (1978) Arch. Biochem. Biophys. 186, 307--316 Greenfield, N. and Fasman, G.D. (1969) Biochemistry 8, 4108--4116 Tung, J.S. and Knight, C.A. (1971) Biochem. Biophys. Res. C o m m u n . 42, 1117--1121 Tanford, C. (1968) Adv. Protein Chem. 23, 121--282 Meyer, M.L. and Kauzmann, W. (1962) Arch. Biochem. Biophys. 99, 348--349 Jirgensons, B. (1963) J. Biol. Chem. 238, 2716--2722 Takagi, T. and Isemura, T. (1960) Bull. Chem. Soc. Japan 33, 437--441 Rao, V.S.R. and Foster, J.F. (1963) Biopolymers 1, 527--544 Tanford, C. (1961) Physical Chemistry of Macromolecules, p. 407--411, John Wiley and Sons, N e w York 22 Hamauzu, Z., Nakatani, N. and Yonezawa, D. (1975) Agr. Biol. Chem. 39, 1407--1410 23 Williams, J.G. and Gratzer, W.B. (1971) J. Chromatogr. 57, 121--125