The effects of diverse proteins on the solubilization of various hydrophobic probes by protein · detergent complexes

219

Biochimica et Biophysica Acta, 534 (1978) 219--227 © Elsevier/North-Holland Biomedical Press

BBA 37927

THE EFFECTS OF DIVERSE PROTEINS ON THE SOLUBILIZATION OF VARIOUS HYDROPHOBIC PROBES BY PROTEIN DETERGENT COMPLEXES •

K U L B I R S. BIRDI a and J A C I N T O S T E I N H A R D T

b

a Fysisk-Kemisk Institut,Technical University,Lyngby, 2800 (Denmark) and b Department of Chemistry, Georgetown University,Washington, D.C., 20057 (U.S.A.) (Received September 28th, 1977)

Summary The solubilization behavior of various protein, detergent complexes with respect to a particular water-insoluble organic substance ("hydrophobic probe") dimethylaminoazobenzene, was reported in earlier studies. The present report describes further the solubilization of other hydrophobic probes (e.g. Sudan II, naphthalene, anthracene and azobenzene) in various protein • sodium dodecyl sulfate complexes, in order to erflarge the scope of our understanding of these phenomena, which undoubtedly play a part in the transport of different water-insoluble organic substances in the living organisms. Solubilization by the various protein • SDS complexes is found to be specific for each probe. The amount of a particular probe solubilized is nearly always equal to the amounts which are solubilized by pure SDS micelles equivalent in amount to the SDS bound. Serious exceptions are found with two heme proteins (e.g. myoglobin and hemoglobin) and a few others. The hemeprotein-SDS complexes also exhibit regions of fiat plateaus in the solubilization curves, whereas the binding equilibria show progressively larger amounts of SDS bound. The solubilization of probes by cationic detergent (cetylpyridinium chloride and cetyltrimethylammonium bromide), protein complexes indicate that the solubilization phenomena are related to the environment of the binding sites (the cationic detergents are known to bind at different sites on the protein than the anionic detergents, i.e. SDS in the present case). With anionic detergents the effective chain length of the pseudo-micellar protein • detergent clusters is sufficient to cause an increase in solubilizing effectiveness of about 1.5 between the complexes and pure micelles. When small probes such as naphthalene are used such ratios are found. With larger probes the effectiveness ratio is reduced to 1.0 or even less as a result of steric interference with the formation of the protein • detergent • probe clusters. The solubilization energy exhibited by each protein • detergent complex is largely determined by the individual protein, and by the charge on the detergent.

220 Introduction The interaction of surfactants, such as sodium dodecyl sulfate, with many proteins, leads to the formation of protein • amphiphile complexes and to the disruption of the tertiary structure of the proteins [1,2]. In recent papers we have reported that water-insoluble organic compounds (henceforth referred to as hydrophobic "probes") such as dimethylaminoazobenzene, which can be dispersed in all detergent micelles, are also solubilized in protein solutions conraining SDS at concentrations at which no SDS micelles are present, i.e. the free SDS concentrations are much lower than the critical micelle concentration. With many proteins, including human serum albumin, bovine serum albumin and ovalbumin, the amount of probe solubilized by the protein • amphiphile complexes is equivalent to the amount of probe which is solubilized by equivalent quantities of pure SDS micelles, i.e. the bound SDS (upto 0.3--0.5 g SDS/protein) behaves as micellar SDS [3,4]. When later studies [4] extended these observations to 13 different proteins, including detailed reversible binding isotherms of SDS with each protein, significant differences in the free energies of binding of SDS by various proteins were found. Thus for example, hemoglobin and myoglobin both have high affinities and high binding capacities, while 7-globulin, apoferritin, and transferrin initially have very low affinities, but change drastically at higher concentrations. The ratio, amount of probe solubilized (absorbance units)/bound SDS(B), as calculated from our published data at free SDS concentration equal to the critical micelle concentration varies as follows: apoferritin (1.2), 7-globulin, conalbumin, bovine serum albumin, human serum albumin, ovalbumin (1.0), 13-1actoglobulin (0.89), adolase (0.83), lysozyme (0.75), catalase (0.64), methaemoglobins (0.6), meth-myoglobin {0.5). In the present paper we report similar data for some of the same proteins and the same detergent with other water-insoluble organic compounds: Sudan II, azobenzene, naphthalene, and anthracene. The solubilities of these probes in the same surfactant micelles differ by large factors. Since the bound surfactant acts similarly to that of pure micelles, it would be expected that the amounts solubilized by the corresponding protein • detergent complexes might vary by the same factor. If they do not, the discrepancies should yield additional information as to the nature of the probe detergent-protein interaction, or about the proteins themselves. Materials and Methods The sources of the proteins used, and of the one of the probes, dimethylaminoazobenzene, have been previously described [4]. The other probes, were as follows: Sudan II from B.D.H., azobenzene from B.D.H., naphthalene from B.D.H. (highest purity grade), anthracene from B.D.H. (highest purity grade), Orange OT from GURR, U.K. The buffers, phosphate 0.033 ionic strength, at either pH 7.1 of 6.4, were the same as used in our earlier papers. The temperatures were 20°C with dimethyiaminoazobenzene and 25°C with the other probes. The equilibrium dialysis method of measuring the binding of SDS with ass radio-active assay, has been described [4]. The solubilizations of the various probes were measured using

221

the same procedure as described earlier by the absorbance m e t h o d [4]. The molar absorptions were determined for each probe after solubilizing known amounts in ethanol or chloroform. The molar absorptions of various probes were at the respective wavelengths of m a x i m u m absorption (the values in parentheses are from literature): dimethylaminoazobenzene (in e t h a n o l ) 2 . 2 . 104 1/mol cm, 410 nm ( 2 . 4 - 1 0 4 1/mol cm, [5]); Sudan II (in chloroform) 1.725 • 104 1/mol cm, 495 nm; naphthalene (in ethanol) 5.12 • 103 1/mol cm, 277 nm; anthracene (in ethanol) 1.67 • l 0 s 1/mol cm, 255 nm; azobenzene (in ethanol) 5.35 • 10: 1/mol cm, 440 nm; Orange OT (in chloroform) 2.77 • 104 1/mol cm, 450 nm ( 1 . 8 . 1 0 4 1/mol cm, [5]). Three sets of measurements were c o n d u c t e d with each probe-protein pair b y methods previously described [4]: (a) Solubility of the probe in micelles of SDS; (b) solubility of the probe in solutions containing protein • surfactant complex and free amphiphile, b u t no amphiphile micelles, and (c) binding of amphiphile to protein, expressed as either g per g or mol per mol. In all cases the surfactant concentration was varied over a wide range, as in earlier work. Results

(A ) SDS micellar solubilization (in the absence o f protein) The measurements reported in this paper are restricted to probes which do n o t detectably effect the free energy of micelle formation, i.e. the critical micelle concentration or the aggregation n u m b e r is n o t afffected b y solubilization [6,7]. In a solubilization experiment the aqueous amphiphile solution is equilibrated while being stirred in c o n t a c t with excess probe in the solid state. The following relations hold between the phases at equilibrium [6] : ~P~obe = ~probe _-- , ~ r o b e SOlid ~" aqueous k~nicelle

(1 )

For these relations between the chemical potentials,/~i, one obtains an expression for the free energy of solubilization of the probe in micelles if ideal behavior m a y be assumed [6] :

AG°~ = --R T ln( c~m/c~ q)

(2)

when AG~° is t h e standard free energy change in transporting 1 mol of probe from the aqueous phase to the micellar phase, and C~q and C~ are the concentrations of the probe in the aqueous phase and in the micellar phase, respectively. Since, in the present study, we compare the energies of solubllization of the same probe in a micellar system with the energy of solubilization in the protein • SDS complex, the term C~q in the above equation is constant and can be included in AG~°' as follows [6] :

AV°~' = - - R T In C~n

(3)

The values of AG,°' for the various probes used in this s t u d y are given in Table I. These values are a measure o f the reluctance of the probes to enter the micellar phase. For example, anthracene is solubilized with a difference in energy of 3929 -- 1557 = 2372 cal/mol (positive) from that of naphthalene. Table I also fists another such measure: the values of mol SDS/mol probe, or 1/C~. Note

222 TABLE I F R E E E N E R G Y O F S O L U B I L I Z A T I O N , AG s L A R S O L U T I O N S O F SDS

0t

, OF T H E V A R I O U S PROBES IN A Q U E O U S MICEL-

C o n d i t i o n s : 25°C, I o n i c s t r e n g t h 0 . 0 3 3 , p H 7.4 Probe

A G s 0t ( c a l / m o l )

m o l S D S / m o l p r o b e ( 1 / C sm )

Dimethylaminoaz obenzene S u d a n II Azobenzene Naphthalene

2586 2026 1557 1557 3929 2277

80 31 14 14 780 47

Anthracene

Phenanthrene

that these ratios are unrelated to the micellar aggregation n u m b e r (89 for our experiments) which depends only on the amphiphile and the temperatrure. (B) Solubilization o f the various probes in protein • S D S complexes Solubilization in bovine serum albumin • S D S complexes. The solubilization

of one of the probes, i.e. dimethylaminoazobenzene in such complexes has been described elsewhere [4]. That data is included here in a somewhat different context. Fig. 1 shows * the a m o u n t o f probe solubilized versus the SDS b o u n d per gram bovine serum albumin (B). Fig. 1 also shows plots of the amounts o f each p r o b e solubilized in pure SDS micelles **. There is no solubilization b y the protein • SDS complex when B is substantially less than a b o u t 0.1 g SDS/g bovine serum albumin, i.e. there is no solubilization b y the first 15--20 molar equivalents of SDS b o u n d per mol of protein. Such differences, whatever their immediate causes, indicate the initiation of conformation changes [1,2]. Attention is invited to the fact that the probe Orange OT, which is solubilized by all micelles (SDS micelles in Fig. 1) is n o t solubilized at all by the bovine serum albumin-SDS complex [8]. The straight line shows that the solubilization per gram of b o u n d SDS is constant in the region u p t o B = 0.5 g. In the case o f dimethylaminoazobenzene, the data [4] established that the results with p r o b e , bovine serum a l b u m i n . SDS and p r o b e " SDS are very similar upto B = 0.5 g. The plot of bovine serum albumin • SDS is also linear for Sudan II although the slope is greater. Azobenzene is also solubilized to a b o u t the same extent in the complex and in SDS micetles. The plots for naphthalene and anthracene show that the complex solubilizes more of both probes than does the pure SDS micelles, although both micelles and complexes have a constant ratio o f solubilization per gram o f SDS bound. In SDS micelles the value of 1/C~ for anthracene is 56 times as large as that for naphthalene {Tabel I). The data in Fig. 1 and Table II show that the amounts solubilized b y the bovine serum albumin • SDS complex also differ b y a b o u t the same factor, i.e. a b o u t 50. Again, Orange OT is n o t solubilized at all b y the complex. * B i n d i n g i s o t h e r m s a r e t a k e n f r o m ref. 4. ** With m o r e p r o t e i n s , v a l u e s a b o v e a b o u t B = 0 . 5 g/g are a t t a i n e d in t h e r e g i o n in w h i c h m i c e l l e s a r e a l s o formed.

223

too,e/,

nV V

mo~ e/{

2.16' .'~" (~ ,

"v•

~

•

v

vV

/

..;.~."

~ --

N

V

,~.'

~ ld'

~ ~ ..;~?" ~ " ~ .~.;." oo ~ ~ ,~:' ~ ..-"

Cl

_oo ...-" ~

.~L" o .--" a ~:~" 0 0....-""

~

~ ./~.~ ~~ ~ , ,

.~

~

~ ~

V V

-6

2~10

V

/¢f/~

~~ /~ ~ ~// ~// / ~/ ~

~

///

~ V

~

~

/// ss

/s s ~l s ~ ~/~

~

~,/~ ~ ,f~

~j~,, ~

~ ~

..~/

.~

"

~ _~' o9-"_~-~'~**~'°'~'"- / ~

~

~

-I --

~d' ~.,d~

n

o e~

~

/i //,

•

.,;;~

v

•

2.1(~"L,,.~. o"e

,i

0

v o

.... ,

I

~-

_

0.2

OA

0.6

0.~

0.8

0.4

0.6

0.~

B

B

BOUND SDS (g SDS/g PROTEIN) BOUND

SDS (gSD~9

PROTEIN)

Fig. I. The dependance of probe solubilization (in tool/l) on amounts of S D S bound (B) by 0.1% solutions o f b o v i n e s e r u m a l b u m i n , d , 4 , O r a n g e O T ; o , e , d i m e t h y l a m i n o a z o b e n z e n e ; D, ), S u d a n I I ; /x, 4 , a z o b e n z e n e ; D, m, n a p h t h a l e n e ; v , v , a n t h r a c e n e . O p e n s y m b o l s : b o v i n e s e r u m a l b u m i n + S D S , c l o s e d s y m b o l s : S D S . T h e a b s c i s s a i n t h e c a s e o f S D S m i c e l l a ~ s o l u t i o n s (i.e. t h e c l o s e d s y m b o l s ) r e f e r s t o c o n c e n t r a t i o n s a b o v e t h e c ~ i t i c a l m i c e l l e c o n c e n t r a t i o n (i.e. B e q u i v a l e n t t o S D S c o n c e n t r a t i o n i n m i c e l l a ~ f o r m ) . F i g . 2. T h e d e p e n d a n c e o f p r o b e s o l u b f l i z a t i o n ( i n m o l / l ) o n a m o u n t s o f S D S b o u n d b y 0 . 1 % s o l u t i o n s o f ~ - l a c t o g i o b u l i n . O p e n s y m b o l s : l a c t o g l o b u l i n + S D S ; c l o s e d s y m b o l s : S D S , a s g i v e n u n d e r F i g . 1.

TABLE

II

AMOUNT OF PROBE SOLUBILIZED (in mol/1) BY VARIOUS PROTEIN • SDS COMPLEXES WITH B ffi 0 . 5 g S D S / g P R O T E I N , A N D T H E R A T I O O F A M O U N T O F P R O B E S O L U B I L I Z E D BY THE COMPLEX TO THAT OF PURE SDS MICELLES EQUIVALENT TO 0.5 g SDS (VALUES IN PARANTHESES)

Protein

Bovine serum albumin

Ovalbumin

Dimethyl-

Sudan II

Azobenzene

Naphthalene

Anthracene

Orange

aminoazobenzene (M r = 225)

(M r = 276)

(M r = 182)

(M r = 128.2)

(M r = 178.2)

OT (Mr = 266)

14.4 • 10 -5 (1.23)

22.3 • 10 -5 (1.87)

2.2 • 10 -5 (1.10)

2.2 • 10 -5 (1.1)

~-Lactogiobulin Hemoglobin

2.34. (1.07)

10 -5

1.3 • 10 -5 (0.6)

Myogiobin

0.8 • 10 -5

(0.37)

7 • 10 -5 (1.26)

4.5 • 10 -5

8.5 ~ 10 -5

20 • 10 -5

(0.81)

(0.73)

(1.67)

4.5. 10 -5 (0.81)

9.2 • 10 -5 (0.79)

16.5 • 10 -s (1.38)

--

--

13 • 10 -$

--

--

(1.09)

---

--

13.5 • 10 -5

--

(1.13)

4.46 • 10 -6 (1.98)

0 --

4.0 • 10 -6

0

(1.77)

--

3.3 • 10 -6 (1.46)

0 --

3.6 • 10 -6 (1.6)

2.7 ' 10 -6

(1.2)

0 --

0 --

224

Solubilization in ovalbumin • SDS complexes. In general the solubility plots (not given) are very similar to those of bovine serum albumin-SDS, as described above, although the absolute magnitudes differ, i.e. upto B = 0.5 g. (Table II}. Orange OT is not solubilized by the complex. Solubilization in ~-lactoglobulin. SDS complexes. Fig. 2 shows that the solubilization of several probes in lactoglobulin • SDS complexes as affected by the amount of SDS bound, B. The probes, dimethylaminoazobenzene and Sudan II, show that the lactoglobulin • SDS complexes behave, relative to pure micelles, just as with bovine serum albumin and ovalbumin. However, the solubility of azobenzene in protein. SDS is less than in the pure SDS micelles. Nevertheless, with naphthalene and anthracene the complex disperses more of each probe than do the micelles, just as with the other proteins described above, i.e. upto B = 0.5 g (Table II). Again there is no solubilization of Orange OT by the protein-SDS complexes. Solubilization in protein-SDS complexes of two heme proteins (whale metmyoglobin and human met-hemoglobin). Results for solubilization of various probes in myoglobin-SDS complex are given in Fig. 3. The non-solubilizing regions correspond to about 5 or 20 equivalents for myoglobin (Fig. 3) and hemoglobin (data not given), respectively (i.e. the same B = 0.1 g SDS/g protein, as for the various protein complexes described so far). The solubility increases abruptly at B = 0.1 g SDS in these heme proteins, thereafter, the solubility remains unchanged in the region B = 0.2--0.5 g SDS. This behavior is observed with all the various probes with both myoglobin and hemoglobin, suggesting a similar solubilization mechanism in both complexes. The SDS added in the plateau region is ineffective in solubilizing and must differ in

//!

mole/I

-4/

2x10 ~4,,0-

~

~

2~

nlv ~W ~

e

w ~ ~

~F /

~

-

~

~ ~

_~,s~"

....

"~/¢/

19/

~ ~6'

o

,f~¢~y

>

=

' ~

,',,~" ~ ,/:~

w

L ~ 0

,1(~s

,,%,/::" , ~/~

~ . ' ~ " ~,5~ "~I"

~

~

~ ~o'~.~ ~ I ~ ~ 0

//i //,/,~

,,:~J

ff n'~;'~/

I

~/

~ A

,~ "

~o

I,K . . . . . . . 0.2

OA

0.6

0.8

1.0

B

BOUND S D S

(gSDS/PROTEIN)

Fig. 3. T h e d e p e n d ~ c e o f p r o b e s o l u b i H z a t i o n ( ~ m o l / l ) o n ~ o u n t s o f S D S b o u n d (B) b y 0 . 1 % solutions o f m y o ~ o b ~ . O p e n s y m b o ~ : m y o ~ o b ~ + S D S , c l o s e d s y m b o l s : S D S , ~ given u n d e r Fig. 1.

225 organization from the solubilizing part. Attention is invited to the fact that the plateau region occurs at SDS concentrations at which binding studies show sharply changing amounts bound and high affinities to SDS. The probe Orange OT is not solubilized by either of the heme protein complexes. Discussion Three sets of differences between micelles and protein, amphiphile complexes require explanation: (A) The fact that the ratios of amounts of hydrophobic probes solubilized by protein • SDS complexes relative to the amounts solubilized by pure SDS micelles (on a basis of equal amounts, 0.5 g of SDS) often differ with the probe used, i.e. the ratios are not always constant (Table H); (B) The fact that with most proteins the ratio of probe dissolved by SDS bound is at a maximum when 0.1 to 0.2--0.3 g/g are bound, and then falls gradually; and (C) That these ratios depend on the protein used. We may expect the protein • SDS complexes, other things being equal, to solubilize more probe than SDS in pure micelles since the "effective chain length" of dodecyl sulfate is less in a micelle than in a protein complex *. It has been shown by one of us [6,7] that the free energy of solubilization of such probes, AG~°', by micelles is linearly related to the length of the alkyl chain of the amphiphile molecule. The enhancement of solubilization corresponding to an increase from 11 to 12 carbon atoms [6] corresponds to a factor of about 1.5. This expectation is qualitatively satisfied by our data, Table II, except in those cases, such as dimethylaminoazobenzene, where the complex seems to solubilize the same amount as the pure SDS micelles. The latter effect may be due to steric effects which reduce the accessibility to the probe of some of the clusters of bound SDS. Many molecular parameters, such as polarity, polarizability, shape, flexibility and molecular weight, may be expected to affect the results. However, if we use the molecular weight as a very rough measure of the probe size, we see that the smallest probe, naphthalene has a very high over-all effectiveness ratio (as does anthracene), and the largest probe Orange OT an effectiveness ratio of zero (Table II). The balance between these factors, and the 1.5 factor advantage of the complexes over the micelles cause their solubilizing properties of dimethylaminoazobenzene and Sudan II to fortuitously resemble those of the pure micelles of SDS. The steric factors that we have referred to may represent merely differences in size distributions of the bound SDS clusters in the protein complexes. As a result of the present work with five probes, and a few proteins (5 out of the 13 already reported), and a single amphiphile, anionic SDS, we have learned that the solubilization of water-insoluble organic compounds (many of which play a part in physiological functions) will often be promoted by the formation of complexes between proteins and amphiphiles. This promotion operates in * T h e c h a r g e o n t h e h e a d g r o u p o f S D S acts t o k e e p its p r o x i m a l C H 2 g r o u p in c o n t a c t w i t h t h e w a t e r at t h e m i c e l l e - w a t a r i n t e r f a c e , t h u s leaving o n l y 11 c a r b o n a t o m s available for t h e h y d ~ o p h o b i c i n t e r a c t i o n s [ 9 ] . In t h e u n f o l d e d p r o t e i n s ( a n d p o s s i b l y in t h e c o m p a c t f o r m a l s o ) , t h e n e g a t i v e charge o n t h e h e a d group o f t h e b o u n d S D S m o l e c u l e is n e u t r a l i z e d b y t h e p r o t e i n s c a t i o n i c c h a r g e , a n d t h u s all t h e 1 2 c a r b o n a t o m s are available for t h e h y d r o p h o b l c i n t e r a c t i o n s .

226 A ,/ • //

A ~,15 n m]

i

s

C PC/'@

1,0

•

]

/• /

© 0

/

©

"I

1°01 b (415rim)

© /"

0.5

0

/ //

CMC

0

/~

~

0.5 ~--

C PC + BSA

BSA ÷

©

©

/~© 0 0,", I

If

I

0.2

(o[4)

(s) I

O6 CPCtotal(g/I)

~

© ,

,

,

lB."

CMC

O

. . . . 0.4

©

$.DS©

~ /

~ O.S

,

, (o..4)~,'~

,

SDS ~

~ (B~

1.0

SD$

total g/L

Fig. 4. (a) The effect of 0 . I bovine serum albumin (BSA) on the solubilization (absorbance, A) of probe dimethylaminoazobenzene by cetylpyr/dinium chloride (CPC) (pH 7.4, Phosphate buffer, ionic strength 0.033, 25°C). e, CPC; o, CPC + BSA. (b) The effect of 0.1% bovine serum albumin (BSA) on the solubilization of dirnethylaminoazobenzene (DMAB) by SDS (pH 7.4, phosphate buffer, ionic strength 0.033, 25°C) e, SDS; o, SDS + BSA. CMC, critical micelle concentration.

two ways: (a) by inducing the formation of the solubilizing clusters at free surfactant concentrations below dritical micelle concentration and by (b) making the complexed surfactant act as if it had a slightly greater hydrocarbon chain length than in the normal micelles. A further conclusion can be drawn, that the energy o f interaction between each SDS molecule and the protein molecule is determined by the environment o f the binding sites. It is reasonable to believe that while such ionic amphiphiles as detergents are primarily attached to the protein molecule by hydrophobic interactions, they tend to cluster about side chains of opposite electric charges. Cationic detergents are reported to bind very strongly to various proteins {refs. 10 and 11, and Birdi, K.S. unpublished). The solubilization of probe dimethylaminoazobenzene by cetylpyridinium chloride • bovine serum albumin complex will be reported elsewhere (Birdi, K.S., unpublished) and is given in Fig. 4a, and for comparison the solubilization of the same probe in $D$" bovine serum albumin complex is given in Fig. 4b. It is clear that the solubilization mechanisms are different in the two systems. The amounts of cetylpyridinium chloride bound are given in Fig. 4a ( as determined from gel filtration chromatography). Comparing the amounts of dimethylaminoazobenzene solubilized by both cetylpyridinium chloride • bovine serum albumin complex and SDS • bovine serum albumin complex at equivalent amphiphile bound (B = 0.4 g amphiphile/g bovine serum albumin), we find the following. The ratio (in absorbance units) for amount dimethylaminoazobenzene solubflized in SDS. bovine serum albumin:SDS (equivalent to 0.4 g as micelles) = 1.0, while this ratio f o r cetylpyridinium chloride, bovine serum albumin:cetylpyridinium chloride (equivalent to 0 . 4 g as micelles)= 0.5. These results support the hypothesis that the amphiphile and the environment of the binding site on the particular protein are the determining factors o f the hydrophobic interactions

227

in the protein • amphiphile molecule complex. If this conclusion is valid, the bovine serum albumin • SDS complex is paradoxically more hydrophobic than the bovine serum albumin, cetylpyridinium chloride (or bovine serum albumin • cetyltrimethylammonium bromide complex), whereas the pure cetylpyridinium chloride or cetyltrimethylammonium bromide micelles are more hydrophobic than the SDS micelles. As described earlier [4], the binding equilibrium involving protein and SDS owes its appearance of high cooperativity (upto B = 0.4 g/g) to the fact that large amounts of amphiphile are bound only after the protein molecule has undergone a drastic unfolding, which exposes many binding sites inaccessible to the ligand before unfolding. The comparative differences between the solubilization energies of different protein. SDS complexes for the same probe might be related to the unfolding differences on SDS binding. This same argument might also be the explanation for the solubilization differences exhibited by bovine serum albumin-SDS and bovine serum albumin, cetylpyridinium chloride (or cetyltrimethylammonium bromide).

Acknowledgements J.S. gratefully acknowledges support of this work by the National Science Foundation (U.S.A.). K.S.B. thanks NATO for the support of expenses of collaboration. The excellent technical assistance of Mrs. J. Klausen is acknowledged. References 1 Steinhardt, J. and Reynolds, J.A. (1969) in Multiple Equilibria in Proteins (Steinhardt, J. and Reynolds, J.A., eds.), Academic Press, New York 2 Steinhardt, J. (1975) in Protein-Ligand Interactions (Sund, H. and Blauer, G., eds.) pp. 412--426, Walter de Gmyter, Berlin 3 Steinhardt, J., Stocker, N., Carroll, D. and Birdi, K.S. (1974) Biochemistry 13, 4461--4468 4 Steinhardt, J., Scott, J.R. and Birdi, K.S. (1977) Biochemistry 16, 718--728 5 Jacobs0 P.J., Geer, R.D. and Anacker, E.W. (1972) J. Colloid Interface Sci. 30, 611~620 6 Birdi, K.S. and Magnusson, T. (1976) Colloid Polymer Sci. 254, 1059--1061 7 Birdi, K.S. (1977) in Micellization, Soinbilization and Microemulsions, ACS Symposium Series (Mittal, K.L., ed.) Vol. 1, pp. 151--169, Plenum Press, New York 8 Birdi, K.S. and Steinha_~it, J. (1975) Fed. Proc. Fed. Am. Soc. Exp. Biol. 34, 598 9 Tanfo~d, C. (1973) in The Hydrophobic Effect (Tanford, C., ed.), pp. 45--59, J. Wiley, New York 10 Birdi, K.S. (1973) Biochem. J. 135, 253--255 11 Few, A.V., Ottewill, R.H. and Parrei~a, H.C. (1955) Biochim. Biophys. Acta 18, 136--137