Lipid Monolayers - Advances in Chemistry (ACS Publications)

---

14 Lipid Monolayers Effect of Phosphatidyl Choline and Cholesterol on the

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

Interaction of Dihydroceramide Lactoside with Rabbit

γ-Globulin 1

GIUSEPPE COLACICCO and MAURICE M. RAPPORT

Department of Biochemistry, Albert Einstein College of Medicine, Yeshiva University, Bronx, Ν. Y. 10461

Discontinuities are seen in the relationship between increase in film pressure,

ΔII, and lipid composition following

the

injection of globulin under monolayers of lecithin—dihydro­ ceramide lactoside and lecithin-cholesterol

mixtures.

breaks occur at 80 mole % C -dihydrocaramide 16

and 50 mole % cholesterol.

The

lactoside

Between 0 and 80 mole %

lactoside and between 0 and 50 mole % cholesterol

the

mixed films behave as pure lecithin. Two possible explana­ tions are: the formation of complexes, having molar ratios of lecithin-lactoside

1 to 4 and lecithin-cholesterol 1 to 1;

and/or

the effect of monolayer configurations (surface mi­

celles).

In this model, lecithin is at the periphery

of the

surface micelle and shields the other lipid from interaction with globulin.

' " p h e participation of ceramide lactosides in immunological reactions (15) prompted interest i n the surface properties of these lipids. Previous work (7) on the nonspecific interaction of monolayers of syn­ thetic dihydrocaramide lactosides with rabbit γ-globulin showed the dependence of the magnitude of the interaction, ΔΠ, on the initial film pressure of the lipid, the length of the fatty acid chain of the sphingolipid, the temperature, and protein concentration. The data were i n accord with earlier experience in lipid-protein interaction (JO, 11, 12) i n showA

1

Present address: New York State Psychiatric Institute, New York, Ν. Y.

157 Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

158

MOLECULAR ASSOCIATION IN BIOLOGICAL AND RELATED SYSTEMS

ing that Δ Π decreased as the initial film pressure increased, Δ Π increased with increasing length of the fatty acid chain, and the curve relating Δ Π to protein concentration had marked discontinuities. A relationship be­ tween lipid structure and interaction with protein was seen. The magni­ tude of Δ Π decreased i n the order cholesterol > Ci -dihydroceramide lactoside > phosphatidyl ethanolamine > phosphatidyl choline > > sphingomyelin. Although glycosphingolipids are the specific lipid components i n the antigen-antibody complex, their activity is markedly enhanced by other (auxiliary) lipids such as lecithin and lecithin-cholesterol mixtures ( 15). The present study deals with the effect of lipid composition on the pene­ tration of lactoside-cholesterol and lactoside-lecithin monolayers by rab­ bit γ-globulin. W e also investigated the lecithin-cholesterol system. Furthemore, since criteria for the existence of lipid-lipid complexes i n monolayers are still few (8, 17), we have used infrared spectroscopy to examine lipid mixtures for the presence of complexes.

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

6

Experimental Materials. LIPIDS AND PROTEIN. The sources of synthetic N-palmitoyl and N-stearoyldihydrosphingosyl lactosides, phosphatidyl choline (egg), cholesterol, and rabbit γ-globulin have been described ( 7 ) . W A T E R AND SALT SOLUTIONS. Distilled water was redistilled over alkaline permanganate from all-glass apparatus. Water of low conduc­ tivity (1 /miho per cm.) was collected i n polyethylene bottles and stored at room temperature ( 2 5 ° C . ) for not longer than 2 days before use. Phosphate buffer ( p H 7.0) containing 0.04M potassium phosphate and 0.10M NaCI was used for the aqueous subphase. It was prepared with low conductivity water and reagent grade salts and stored i n poly­ ethylene bottles at room temperature for not longer than 5 days. Protein solutions (1 mg. per ml.) were prepared with this buffer and stored i n borosilicate glass flasks for not longer than 5 days at 5 ° C . ORGANIC SOLVENTS AND LIPID SOLUTIONS.

Reagent grade solvents

were distilled before use. The lipids were dissolved in chloroform-methanol (85 to 15 v . / v . ) at a concentration of about 0.5 mg. per ml. and stored i n glass-stoppered borosilicate glass tubes for not longer than 5 days at 5 ° C . L i p i d mixtures were freshly prepared from the stock solu­ tions of the individual components. Apparatus and Procedure. SURFACE ISOTHERMS. The technique for determining the Π - Α and AV-A curves of the lipid films has been de­ scribed (6). Briefly, the Wilhelmy plate method was used to measure surface tension, from which the surface pressure was calculated ( Π = 7H o-yfiim). The surface potential was measured by means of a radio­ active ( R a ) air electrode and a saturated calomel electrode connected to a high impedance model 610 Β Keithley electrometer ( Keithley Instru­ ments, Cleveland, O h i o ) . LIPID-PROTEIN INTERACTION. The technique of Doty and Schulman was used to bring lipid monolayer and protein into contact ( 7 ) . A rec2

2 2 6

Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

14.

Lipid Morudayers

COLACICCO AND RAPPORT

159

tangular Lucite trough (25 X 7.6 cm. and 1.9 cm. deep) was filled to the rim with the aqueous subphase at 25 °C. The surface was cleaned, and the lipid was spread. The film was then compressed until the pres­ sure was 2 dynes per cm., the total area being about 80 sq. cm. Stirring of the subphase was started, and after 1 minute the protein was injected deep into the subphase. The fall in surface tension (increase in surface pressure) was recorded at 1-minute intervals for 40 minutes. The subphase under the monolayer was stirred by a Teflon-covered magnetic bar that was moved back and forth over the length of the trough by a motor-driven mechanism. The fastest rate which d i d not cause disruption of the monolayer was about 4 strokes per minute. INFRARED SPECTRA. A Perkin-Elmer model 237 spectrometer was used. Solutions of 5 to 10 mg. per ml. of lipid in spectral grade chloroform were placed in a NaCI microcell of 1.0-mm. path length for study i n the region from 2.5 to 6.0 microns. The film technique was used for observa­ tions between 5 and 15 microns, with a beam condenser and attenuator (Perkin-Elmer, Norwalk, Conn.). The lipid, 75 to 100 jmgrams i n either chloroform or chloroform-methanol 85 to 15, was deposited on 1 sq. cm. of the NaCI plate, and the solvent was removed by evaporation under an infrared lamp for 10 minutes. Results Lipid-Protein Interaction. Although measurements were made over a 40-minute period, for convenience in the presentation of the results ( 7 ), only the rise in pressure at 15 minutes is given. A linear relationship was found between film penetration, ΔΠ, and lipid composition for mixtures of cholesterol and either N-palmitoyl or N-stearoyldihydrosphingosyl lac­ toside (Figure 1). In contrast, marked discontinuities were observed with mixtures containing phosphatidyl choline (Figure 2 ) . A 20 mole % concentration of lecithin in Ci -dihydroceramide lactoside or 25 mole % in Cis-dihydroceramide lactoside lowered the value of Δ Π to that for pure lecithin. 6

Similarly, a marked discontinuity was found at 50 mole % in the curve relating Δ Π to the composition of cholesterol-lecithin films ( Figure 3). U p to 50 mole % cholesterol the magnitude of Δ Π was the same as with lecithin alone; above this, Δ Π increased linearly with cholesterol concentration. Infrared Spectra. The effect of composition on infrared absorption of cholesterol-lecithin mixtures was striking i n two regions of the spec­ trum—near 2.8 and 9.2 microns—which are associated with O - H and P-O-C stretching frequencies, respectively (Figure 4 ) . Pure cholesterol in C H C 1 solution has a sharp peak at 2.8 microns, owing to the free O - H group. Lecithin has a broad absorption between 2.9 and 3.2 microns, attributed to bound water (2). The peak at 2.8 microns disappeared in mixtures containing 50 mole % or more of 3

Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

160

MOLECULAR ASSOCIATION IN BIOLOGICAL AND RELATED SYSTEMS

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

lecithin. This indicates that i n these mixtures the cholesterol is probably bound to lecithin in an equimolar complex. The absorption at 2.8 microns was shifted to 2.9 microns, producing a shoulder on the broad "water peak" of lecithin. The intensity of the absorption at 2.8 microns of pure cholesterol solutions increased with concentration but without any shift. This indicates that neither O H - O H interactions of cholesterol nor pos­ sible traces of water i n the chloroform could account for the shift ob­ served with lecithin-cholesterol mixtures.

I Lactoside 100 Cholesterol

0

i

l

l

]

75

50

25

0

25

50

75

100

MOLE % Figure 1. Effect of lipid composition on surface pressure of films in interaction with rabbit y-globulin at 1 ^gram/ml. Ο Cholesterol-C u-dihydroceramide lactoside φ Cholesterol-Cis-dihydroceramide lactoside Measurement at 15 minutes Initial film pressure, 2 dynes/cm. 0.04M phosphate buffer, 0.1M NaCI, pH 7.0, 25°C. The infrared spectrum of lecithin was affected markedly in the region of the P - O - C absorption. The peaks between 9.1 and 9.4 microns showed a continuous shift from the higher frequency to the lower frequency component as the concentration of cholesterol increased from 0 to 75 mole % . The two peaks were equal at a composition of 50 mole % (Figure 4 ) . The infrared spectra of systems containing lactoside could not be analyzed i n a similar way because the absorption of the hydroxyl groups of the sugar masked the absorptions of cholesterol at 2.8 microns and of the P - O - C group of lecithin between 8 and 10 microns.

Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

14.

COLACICCO AND RAPPORT

I Lecithin

100

Lactoside 0

161

Lipid Monolayers

IL-j I ι

I

ι 1

20

10

0

80

90

100

MOLE %

Figure 2. Effect of lipid composition on sur­ face pressure of films in interaction with rabbit γ-globulin at 1 ^gram/ml. Ο Egg lecithin-Cle-dihydroceramide lactoside φ Egg lecithin-Ci8-dihydroceramide lactoside Measurement at 15 minutes Initial film pressure, 2 dynes/cm. 0.04U phosphate buffer, 0.1M NaCI, pH 7.0, 25°C. Discussion A remarkable effect of lecithin is seen i n the discontinuities of the ΔΠ curves for both the lecithin-] actoside and the lecithin-cholesterol systems (Figures 2 and 3) when these are compared with the linear in­ crease of Δ Π i n the cholesterol-lactoside system (Figure 1). Since these effects probably result from the organization of lipids in the monolayer, we must first consider criteria for ascertaining the nature of lipid-lipid associations. Moreover, since the discontinuities i n the ΔΠ-composition curves probably reflect transitions i n monolayer con­ figurations, we discuss these from the standpoint of penetration of protein into the lipid film and its relation to specific lactoside-antibody interaction. Lipid-Lipid Associations Three criteria for lipid-lipid interactions i n lipid mixtures are the mean area of the lipids in the mixed film, the surface potential of the mixed monolayer, and the infrared spectra of the lipid mixtures.

Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

162

MOLECULAR ASSOCIATION IN BIOLOGICAL AND RELATED SYSTEMS

Lecithin Cholesterol

100

75

50

25

0

0

25

50

75

100

MOLE % Figure 3. Effect of lipid composition on surface pressure of films of egg lecithin-cholesterol in inter­ action with rabbit γ-globulin at 1 ^gram/ ml. Measurement at 15 minutes Initial film pressure, 2 dynes I cm. 0.04U phosphate buffer, 0.1M NaCI, pH 7.0, 25°C. Mean Area. If the association of two lipids i n the monolayer is ac­ companied by a reduction i n the average molecular area (film contrac­ tion), the curve relating mean area to composition w i l l show a deviation from molecular additivity (ideal miscibility). The significance of such contractions is still not clear (8, 17), and contraction i n itself is not a sufficient criterion for detecting interaction. O n the one hand, the cholesterol-lactoside system did not show film contraction ( unpublished data ). This was expected since the monolayers of dihydroceramide lactosides ( 7 ) and of cholesterol are not compressible. O n the other hand, lecithin-lactoside and lecithin-cholesterol systems did show contraction, which could have been predicted since the lecithin monolayer is of the expanded type and is very compressible. (The area per molecule of lecithin at 2 dynes per cm. is large, 110 sq. Α., as op­ posed to 52 sq. A . for Ci -dihydroceramide lactoside and 40 sq. A . for cholesterol. ) However, no direct correlation exists between contraction of the mixed film and discontinuities in the ΔΠ-composition curve. The sharp breaks at 80 mole % Ci -dihydroceramide lactoside (Figure 2) and at 50 mole % cholesterol ( Figure 3 ) are not seen in the mean area-compo­ sition curves ( unpublished data ). Since the Δ Π value is constant between 6

6

Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

14.

Lipid Monolayers

COLACICCO AND RAPPORT

163

20 and 100 mole % lecithin with lactoside (Figure 2) and between 50 and 100 mole % lecithin with cholesterol (Figure 3), lecithin probably forms a complex with the other lipid. Surface Potential. Shah and Schulman have proposed that interac­ tion between dipoles of uncharged lipids in mixed monolayers should result in a change in surface potential, AV. Linearity of the relation of AV to composition of the lecithin-cholesterol monolayer was taken to indicate absence of interaction (17). We do not agree with Shah and Schulman, since surface potential does not appear to be a valid criterion for assaying interaction between dipoles of uncharged lipids. Except for the speculations of Shah and Schulman (17, 18), there is neither theo­ retical nor experimental evidence that dipole-dipole interactions have SOLUTION C

0

L

;

3.0 3.5 1 11 1 1 1 j 1 1 1 1 j

FILM 9.0 10.0

2.5

j

\J

100

100

0

40

60

50

50

60

40

75

25

i

X X X . \J X 1 3.0

1 1 1

I I I I

3.5

1

ο

1.

ιυ · ;!

WAVELENGTH

Figure 4.

ι ζ

I

11 2.5

1

9.0 11111111lo.o • 11 μ

Infrared absorption spectra of cholesterol (C)lecithin(L) mixtures

Solution in chloroform, 5 to 10 mg./ml., 25°C; path length, 1 mm. Film, 75 to 100 μgrams lipid on 1 sq. cm. of NaCI plate 2 mm. thick

Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

164

MOLECULAR ASSOCIATION IN BIOLOGICAL AND RELATED SYSTEMS

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

any direct effect on the surface potential. These interactions should not be confused with charge-charge interactions. Therefore, neither mean area nor surface potential changes are a valid criterion for lipid-lipid interaction. Although large deviations from ideality were seen in the mean area—composition curves and little or no deviations were seen in the AV-composition curves, these curves d i d not show the sharp discontinuities present in the ΔΠ-composition curves. However, the absence of such discontinuities indicates neither the pres­ ence nor the absence of molecular interactions. Infrared Spectra. The shift of the absorption of the O H group of cholesterol to lower frequency (from 2.8 to 2.9 microns) in the presence of 50 mole % or more of lecithin indicates that cholesterol is involved in an equimolar complex with lecithin, probably through hydrogen bonding. However, the H acceptor is still not known. Although threedimensional models support the view that the oxygen of the P - O - C group of phosphatidylcholine is a possible H bond acceptor from the O H group of cholesterol, the interpretation of the infrared absorption data is com­ plicated by the fact that the P - O - C peak (9.1 to 9.4 microns) changes continuously from 0 to 75 mole % cholesterol, whereas the shift of the O H peak is no longer discernible above 50 mole % cholesterol (Figure 4). The infrared evidence for hydrogen bonding between cholesterol and lecithin in chloroform solution is no evidence of a similar complex in the monolayer but suggests such a possibility. It does not exclude the hydrophobic bonding suggested by Chapman from N M R studies of the aqueous suspensions of equimolar mixtures of cholesterol and lecithin (3). Monolayer

Configurations

F i l m penetration studies show unequivocally that lecithin-cholesterol mixtures containing from 0 to 50 mole % cholesterol and lecithin-lactoside mixtures containing from 0 to 80 mole % Ci -dihydroceramide lactoside have the same effect as pure lecithin. This suggests the presence of a lipid complex in which lecithin prevents the interaction of the cholesterol or ceramide lactoside with globulin. Over these ranges of composition the lipid film would consist of a mixture of the lecithin-cholesterol or the lecithin-lactoside complex with excess lecithin. One may picture two models in which the protein contact is restricted to molecules of lecithin. In one, individual polar groups of the protein interact with the excess lecithin molecules as well as with the lecithin portions of the complex. In the other model, the protein as a whole interacts with the lecithin sites of polymeric lipid structures. The latter, which could be referred to as surface micelles ( I ) , are visualized also through the term "mono6

Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

14.

c o L A C i c c o AND RAPPORT

165

Lipid Monolayers

layer configurations." The two terms are not interchangeable. Whereas the surface micelle is a physical entity, monolayer configurations are meant to convey the concepts of orientation of molecules and topography of the monolayer.

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

Intermolecular Area and Film Compressibility.

Since the

observa­

tions on film penetration, ΔΠ, serve to define discontinuities in film struc­ ture, the relation of Δ Π to intermolecular areas and film compressibility (13, 16) requires some analysis. Does protein penetrate the interface simply because the lipid monolayer is compressible and has available intermolecular areas, or does the protein interact with the lipid? Schulman (16) and Eley and Miller (12) pointed out that, with monolayers which are incompressible and thus contain little free area, a small quantity of protein is sufficient to cause a large rise in pressure (such as that obtained with cholesterol). The phenomena of intermolecular area and film compressibility do not take into account the mechanism of lipid-protein interaction, nor do they offer a satisfactory explanation for the most recent observations. When applied to the interaction of lipid mixtures with globulin, they lead to an absurdity. To have the same value of Δ Π at 80 mole % lactoside, 50 mole % cholesterol, and 100 mole % lecithin would require the same intermolecular area for films having these compositions. This is not true. Moreover, on the basis of intermolecular area and compressibility one would have expected the magnitude of Δ Π values to be in the order: cholesterol > spingomyelin = hydrogenated egg lecithin = egg lecithin. However, i n the interaction of these lipids with rabbit γ-globulin the values of Δ Π were in the order: cholesterol > hydrogenated egg lecithin = egg lecithin > > sphingomyelin. A n alternative explanation for these data may be found by consider­ ing monolayer configurations. In a suitable model, the protein as a whole would penetrate between the surface micelles. T w o new parameters should thus become important in studying film penetration: intermicellar area, and structure and orientation of lipid molecules at the periphery of the surface micelles. State of Protein in Lipid-Protein Monolayer.

According to

Doty

and Schulman (9), Matalon and Schulman (14), and Eley and Hedge (10, 11, 13), protein coming into contact with the lipid film unfolds, and its hydrophobic residues penetrate between the hydrophobic chains of the lipid; the polar groups of the lipid interact with the peptide groups of the protein. The salient feature of this model is the uniform distribution (interpénétration) of the side chains of the protein between the hydrophobic chains of the lipid. This representation is not supported by two observations. One is that when either trypsin or pronase was injected into the subphase con-

Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

166

MOLECULAR ASSOCIATION IN BIOLOGICAL AND RELATED SYSTEMS

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

taining protein at equilibrium with the lipid monolayer, no fall in surface pressure was observed although hydrolysis of the protein in the subphàse was extensive. If the protein, after penetration in the monolayer, were in the form of extended polypeptide chains, the peptide bonds would be more susceptible to attack by proteolytic enzymes. The release of soluble peptides from the film into the subphase should have caused a decrease in surface pressure, but this was not the case. The absence of appreciable hydrolysis of the protein in the film could suggest that, in entering regions of low dielectric constant at the air-water interface, the protein changes to a more compact structure and is not accessible to the enzyme (4, 5). The second observation which does not support the unfolded protein model is that when phospholipase A (IV. naja venom) was injected into the subphase under the lipid monolayer at equilibrium with globulin, lecithin was readily attacked, as indicated by the rapid fall of surface potential (4, 5, 6 ) . If the penetrated protein were to cover entirely the polar groups of the lipid facing the aqueous subphase ( as postulated in the unfolded protein model), the lipid molecules should not be accessible to the lipolytic enzyme. A model that is consistent with these observations of the action of trypsin and phospholipase A and with the discontinuities in the ΔΠ-composition curves ( Figures 2 and 3 ) is one i n which the lipid monolayer is not a continuous palisade of uniformly oriented lipid molecules but rather an assembly of surface micelles. In this model, proposed by Colacicco (4, 5), the protein first comes into contact with the lipid molecules at the pe­ riphery of the surface micelles and then inserts itself as a unit between them. This is the basis for the generalized nonspecific interaction between lipids and proteins which results in increase of surface pressure. One may thus explain the identical Δ Π values obtained with films of lecithin and 80 mole % lactoside by picturing the lecithin molecules outside and the lactoside molecules inside the surface micelles. In this model lecithin prevents the bound lactoside from interacting nonspecifically with globu­ lin and produces the same increase i n pressure as with a film of pure lecithin. In the mixed micelle the lactose moiety of the lactoside pro­ trudes into the aqueous subphase. Contact of the protein with these or other nonperipheral regions of the surface micelle would not increase the surface pressure. W i t h this model one might predict that when the lactose moiety is responsible for a specific interaction with protein (as i n interaction with antibody directed against lactose), binding of the specific protein to the lactose would block its further penetration into the intermicellar (interlecithin) spaces. Experiments in which a specific antilactoside antibody was used in place of globulin (unpublished data) showed that this was indeed the case. The Δ Π value was much smaller than with nonspecific

Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

14.

COLACICCO AND RAPPORT

Lipid Monolayers

167

globulin. The large value of Δ Π characteristic of the nonspecific inter­ action was restored when the specific interaction was inhibited by the presence of 0.1M lactose.

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

Conclusions The role of lecithin as an auxiliary lipid i n the specific interaction of lactosides with globulin i n monolayers is related to two processes: com­ plex formation between 3 or 4 molecules of lactoside and each lecithin molecule, and the protection of the lactoside molecules i n surface micelles from nonspecific interaction. T h e location of lecithin at the periphery of the surface micelle would explain why the mixed micelle behaves as lecithin i n nonspecific interaction. Lactoside molecules, located i n the center of the surface micelle, would be in a position to interact specifically with antibody i n the aqueous subphase ( 5 ) . The nature of the lecithin-cholesterol association is probably also that of a complex (1 to 1), i n which the O H group of cholesterol is involved through hydrogen bonding. It is not known, however, whether the monolayer consists of a uniform population of bimolecular complexes or of configurations ( surface micelles ) i n which cholesterol molecules are surrounded by an equal number of lecithin molecules. The concept of monolayer configurations is particularly useful i n explaining penetration of the protein as a unit, i n a form i n which it has more structure than i n solution. Studies of interactions between protein and monolayers of mixed lipids are very useful for examining the nature of lipid-lipid associations. Literature (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)

Cited

Abood, L. G., Rushmer, D. S., ADVAN. CHEM. SER. 84, 169 (1968). Chapman, D., "The Structure of Lipids," Wiley, New York, 1965. Chapman, D., Penkett, S. Α., Nature 211, 1304 (1966). Colacicco, G., J. Am. Oil Chemists' Soc., in press. Colacicco, G., J. Colloid Interface Sci., in press. Colacicco, G., Rapport, M. M., J. Lipid Res. 7, 258 (1966). Colacicco, G., Rapport, M. M., Shapiro, D., J. Colloid Interface Sci. 25, 5 (1967). Demel, R. Α., Van Deenen, L. L. M., Pethica, Β. Α., Biochim. Biophys. Acta 135, 11 (1967). Doty, P., Schulman, J. H., Discussions Faraday Soc. 6, 21 (1949). Eley, D. D., Hedge, D. G., Discussions Faraday Soc. 21, 221 (1956). Eley, D. D., Hedge, D. G., J. Colloid Sci. 11, 445 (1956). Eley, D. D., Miller, G., Proc. Intern. Congr. Surface Activity, 3rd, Cologne 1960, II, Β157. Eley, D. D., Hedge, D. G., J. Colloid Sci. 12, 419 (1957). Matalon, R., Schulman, J. H., Discussions Faraday Soc. 6, 27 (1949). Rapport, M. M., Res. Publ. Assoc. Nervous Mental Diseases 40, 159 (1962).

Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

168

MOLECULAR ASSOCIATION IN BIOLOGICAL AND RELATED SYSTEMS

(16) Schulman, J. H . , Discussions Faraday Soc. 21, 272 (1956). (17) Shah, D., Schulman, J. H . , J. Lipid Res. 8, 215 (1967). (18) Ibid., 6, 341 (1965).

Downloaded by UNIV OF ROCHESTER on June 12, 2018 | https://pubs.acs.org Publication Date: June 1, 1968 | doi: 10.1021/ba-1968-0084.ch014

RECEIVED July 17, 1967. Work supported by research grants from the American Cancer Society (P-410) and the National Science Foundation (GB-5984).

Goddard; Molecular Association in Biological and Related Systems Advances in Chemistry; American Chemical Society: Washington, DC, 1968.