Monolayers of sterols and phosphatidylcholines containing a 20-carbon chain

Chemistry and Physics of Lipids, 22 (1978) 207-220 © Elsevier/NorthMolland Scientific Publishers Ltd.

MONOLAYERS OF STEROLS AND PHOSPHATIDYLCHOLINES CONTAINING A 20-CARBON CHAIN R.W. EVANS and J. TINOCO*

Department of Nutritional Sciences, University of California, Berkeley, Calif. 94720 (USA) Received December 13th, 1977

accepted March 9th, 1978

Pressure-area curves of monolayer films were measured for phosphatidylcholines (PC) in which the 1-position was occupied by palmitic acid and the 2-positions were occupied respectively by: 20 : 0, 20 : ln9, 20 : 2n6, 20 : 3n3, 20 : 3n6, 20 : 3n9, 20 : 4n6 or 20 : 5n3 fatty acids. The interactions of these PC with cholesterol or desmosterol were studied. Fully saturated PC (16 : 0-20 : 0) displayed a relatively small molecular area. The presence of one double bond greatly increased the molecular area, but a second double bond resulted in only a small additional increase in area. A third double bond caused a further large expansion in area, but the presence of a fourth or fifth double bond had little additional effect. Condensation of molecular area was observed with all sterol/PC mixed films. Approximately equimolar mixtures of sterols and unsaturated PC condensed maximally, but 16 : 0-20 : 0 PC condensed most in mixtures containing 20-30 mol% of either sterol. The extent of condensation varied with surface pressure. The pressure at which maximum condensation occurred depended upon the structure of the PC and was always 20 dyn/cm or lower. The pressure at which maximum condensation with cholesterol occurred increased with increasing unsaturation of the PC.

I. I n t r o d u c t i o n

The phospholipids of a biological membrane are a complex mixture containing a variety of polar head groups and an assortment o f fatty acid side chains differing in chain length and unsaturation. Because of their location in biological surfaces, the surface properties of phospholipids have been studied in monolayers at the air-water interface. Studies of individual synthetic phospholipids have shown that the structures of the acyl chains greatly affect the behaviour of these molecules in monolayers and also influence the monolayer interactions between phospholipids and cholesterol [1~]. The influence of double bond position has not been systemically investigated in monolayers of phospholipids, although such effects should be clearly measurable. Pressure-area curves reported for each possible positional isomer of methyl.branched palmitic and stearic acids show that the influence of a methyl side chain is greatest when the methyl group is near the center of the chain [3,4]. In these series, the greatest molecular areas occurred in 8-methyl palmitic acid and in 9-methyl stearic * To whom correspondence should be addressed. 207

208

R.I¢. Evans, J. I~noco, Sterols and phosphatidylcholines in monolayers

acid, respectively. Similarly, differential thermal analysis of 16 hydrated phosphatidylcholines (PC), each containing a different positional isomer of cis-octadecenoic acid, showed that the influence of the double bond was greater near the center of the chain. In this series, the limiting transition temperature was -22°C when the double bond was at A-9, but increased to over 40°C when the double bond was moved to either end of the chain [5]. PC molecules containing a variety of 20-carbon fatty acids in the 2-position are attractive experimental molecules because natural phospholipids contain a relatively large number of physiologically important C2o fatty acids. Arachidonic acid, 20 : 4n6, is a normal constituent of PCs from many tissues. Icosapentaenoic acid, 20 : 5n3, occurs in choline phosphatides of fish [6-8] and of rats fed flaxseed oil [9]. Icosenoic acid, 20 : ln9, occurs in small proportions in brain phospholipids [10]. Icosatrienoic acid, 20 : 3n9, is an important component of phospholipids in essential fatty acid (EFA)-deficient animals [11-13] and icosatrienoic acid, 20 : 3n6, is a precursor of certain prostaglandins as well as of 20 : 4n6. In addition, comparison of the isomeric icosatrienoic acids, 20 : 3n3, 20 : 3n6 and 20 : 3n9, would indicate the influence of double bond position. Icosanoic acid, 20 : 0, and icosadienoic acid 20 2n6, were included for comparison, although they are usually very minor compon ents of naturally-occurring phospholipids. Cholesterol is quantitatively the most important sterol in animal membranes, but its immediate biosynthetic precursor, desmosterol, accumulates in certain tissues. Desmosterol differs from cholesterol only by the presence of a double bond at C24 of the side chain. It occurs in high levels in hamster epididymal spermatozoa [14] and in developing human brain [15,16]. Desmosterol also accumulates when rats are treated with 20,25-diazacholesterol, which also produces myotonia and increased levels of polyunsaturated fatty acids in muscle phospholipids [17-19]. In both brain [20] and spermatozoa [21 ], high proportions of polyunsaturation normally occur in phospholipids. This association in vivo of highly polyunsaturated phospholipids with unusually high concentrations of desmosterol prompted us to study the interaction of desmosterol as well as of tholesterol with a series of PC.

H. Experimental A. Gas-liquid chromatography {GLC)

Methyl esters were prepared from free fatty acids and phosphatidylcholines by heating in redistiUed reagent grade methanol acidified with H2SO4 [22] plus, if appropriate, a known amount of heptadecanoic acid as an internal standard (Eastman Organic Chemicals, Rochester, NY). The chromatograph (Model 2100, Varian Aerograph, Palo Alto, Calif.) had a flame ionisation detector and a 1.83 m X 2 mm i.d. column of 10% SP-216-PS on 100-120 mesh Supelcoport (Supelco, Inc., Supelco Park, Bellefonte, Penn.), operated at 165°C. Injection port and detector were maintained at

R. I¢. Evans, J. l~noco, Sterols and phosphatidylcholines in monolayers

209

200°C. Methyl esters of a mixture of free fatty acids(16 : 0, 18 : 0, 18 : ln9, 18 : ln9, 18 : 2n6, 18 : 3n3, 20 : 4n6 and 22 : 6n3) of known weight were prepared and relative sensitivity factors were determined for each methyl ester (16 : 0 = 1.00). For preparative GLC of methyl esters, the chromatograph (Model 810, F & M Scientific Corp., Avondale, Pa.)had a flame ionisation detector with a 10 : 1 splitter, a Tri-Carb GLC fraction collector (Model 830, Packard Inst. Co. Inc., La Grange, I11.) and a 1.83 m X 4 mm i.d. column of 10% SP-216-PS on 100-120 mesh Supelcoport. Oven temperature was 180°C and the injection port and detector were maintained at

2o0°c. Free sterols were chromatographed on a 1.83 m × 2 mm i.d. column of 3% SP2250 on 100-120 mesh Supelcoport (Supelco, Inc.) operated at 275°C. Injection port and detector were maintained at 300°C. B. Phosphorus determination The procedure ofChen et al. [23] was used as modified by Snider and Clegg [24] except that colour was developed for 15 min at 50°C. The samples were then allowed to stand at room temperature for 90 min before reading at 820 nm against a reagent blank. C. Preparation o f lipids Cholesterol (Nutritional Biochemicals Corp., Cleveland, Ohio), recrystallized twice from absolute ethanol and from petroleum ether (B.R. 40-60°C), and desmosterol (Steraloids, Inc., Pawling, N.Y.) each gave only one component on GLC. The sterols were dried over phosphorus pentoxide under vacuum at room temperature overnight, and stock solutions in chloroform were prepared by weight. These solutions were stored at -16°C under nitrogen and used within three days of preparation. Fatty acids: 20 : 0, 20 : ln9, 20 : 2n6, 20 : 3n3, 20 : 3n6 and 20 : 4n6 (Nuchek Prep. Inc., Elysian, Minn.) were over 98% pure as shown by GLC. Seventy percent 20 : 5n3 fatty acid was obtained from Nuchek and purified by preparative GLC of its methyl ester. The final product was > 97% pure. Phospholipids enriched in 20 : 3n9 were extracted from the hearts and livers of EFA-deficient rats and the 20 : 3n9 was isolated by preparative GLC of its methyl ester. The final sample was > 93% pure and contained about 3% 20 : 4n6 and about 3% of what were believed to be other isomers of 20 : 3. Methyl esters of 20 : 3n9 and 20 : 5n3 were converted to free fatty acids by refluxing in 4 ml of 6% KOH in 95% ethanol for 30 min. After cooling, 5 ml of water and a drop of methyl orange were added and the sample was acidified with H2SO4. The free fatty acids were extracted three times with 10 ml of redistiUed petroleum ether (B.R. 40-60°C). The starting material for the synthesis of all the 1-palmitoylphosphatidylcholines was 1,2-dipalmitoylphosphatidylcholine (L-a-dipalmitoyllecithin, Calbiochem, Los

210

R.W. Evans, J. Tinoco, Sterols and phosphatidylcholines in monolayers

Angeles, Calif.). It ran as one component during thin-layer chromatography (TLC) [25] and GLC o f its side chains as methyl esters gave > 99% palmitate. It was treated with Crotalus adamanteus venom (Sigma Chemical Co., St. Louis, Mo.) according to a modified procedure o f Long and Penny [26] to form 1-palmitoylglycerophosphorylcholine. Fatty acid anhydrides were prepared according to the procedure of Selinger and Lapidot [27]. Freshly prepared 1-palmitoylglycerophosphorylcholine was reacylated with anhydride according to the procedure of Cubero Robles and Van den Berg [28]. All preparations were purified by preparative TLC [25]. Table 1 lists the fatty acid compositions o f the synthesized phosphatidylcholines. All PCs were stored at -16°C in chloroform methanol under nitrogen and used within two days o f preparation. All PCs were treated with snake venom (as above) and the lysophosphatidylcholines produced were isolated by TLC [25 ]. The fatty acid contents of the lysophosphatidylcholines are presented in table 2. All the phosphatidylcholines appeared to consist of two isomers. The amount o f inverted isomer (unsaturated fatty acid in the 1-position) ranged from about 10% in 16 : 0 - 2 0 : 4n6 PC to about 19% in 16 : 0 - 2 0 : 2n6 PC. Previous reports indicate that the positions o f the fatty acids in phosphatidylcholines Table 1 Names and fatty acid composition of PCs. Systematic [29] names of PC

Abbreviations a

Mol% Fatty acid 1

Fatty acid 2

16 : 0-20 : 0 PCb

49.8

50.2

16 : 0-20 : ln9 PC

49.3

50.7

16 : 0-20 : 2n6 PC

49.4

50.6

16 : 0-20 : 3n3 PC

53.2

46.8

16 : 0-20 : 3n6 PC

51.6

48.4

16 : 0-20 : 3n9 PC

50.4

49.6 c

16 : 0-20 : 4n6 PC

49.7

50.3

16 : 0-20 : 5n3 PC

53.1

46.9

I -hexadecanoyl-2-icosanoyl

glycero-3-phosphocholine 1-hexadecanoyl-2-icosa- 11-enoyl glycero-3-phosphoeholine 1-hexadecanoyl-2-icosa-11,14-dienoyl glycero-3-phosphocholine 1-hexadecanoyl-2-icosa-11,14,17trienoyl glycero-3-phosphocholine 1-hexadecanoyl-2-icosa-8,11,14trienoyl glycero-3-phosphocholine 1-hexadecanoyl-2-icosa-5,8,11trienoyl glycero-3-phosphocholine 1-hexadecanoyl-2-icosa-5 ,8 ,11,14tetraenoyl glycero-3-phosphocholine 1-hexadeeanoyl-2-icosa-5,8,11,14,17pentaeaoyl glycero-3-phosphocholine

aThe number following 'n' refers to the position of the double bond closest to the methyl end of the chain. b Average of two preparations. CApprox. 6% not 20 : 3n9 (see text).

R. I¢. Evans, J. l~noco, Sterols and phosphatidylcholines in monolayers

211

Table 2 Fatty acid composition of lysophosphatidylcholinesobtained from synthetic PCs treated with Crotalus adaman teus venom Phosphatidytcholine

16 16 16 16 16 16 16 16

: 0-20 : 0-20 : 0-20 : 0-20 : 0-20 : 0-20 : 0-20 : 0-20

: 0a : ln9 : 2n6 : 3n3 : 3n6 : 3n9 : 4n6 : 5n3

Mol% Palmitic acid

20-Carbon acid

81.2 86.9 80.6 85.8 88.7 86.4 89.7 88.9

18.8 13.1 19.4 14.2 11.3 13.6 10.3 11.1

aAverage of two preparations. can be interchanged with relatively little effect on their areas in monolayers or on their monolayer interactions with cholesterol [30,31 ]. D. Pressure~trea m easurem en ts

Pressure-area measurements were made at 22 ° -+ 2°C in a surface balance (Cenco Hydrophil Balance, Central Scientific Co., Chicago, IU.) as described earlier [32]. The temperature for one complete set o f measurements, run on the same day, did n o t vary more than _+0.5°C. The subphase was glass-distilled water, pH 5.1. Before each measurement the water surface was cleaned by repeated sweeps with a polyethylene bar. The appropriate proportions o f sterol and phosphatidylcholine were mixed, the solvent (chloroform/methanol) was removed under nitrogen and the sample was redissolved in a known volume o f application solvent (benzene/chloroform/methanol, 90 : 10 : 6, by vol.). The samples were applied as 50/al aliquots and the measurements, started 2 min later, took 1 0 - 1 5 min. A complete set of mixtures and pure components were measured within one day and at least two films o f each composition were spread. Duplicate aliquots o f every sample (except pure sterol) were removed at time o f spreading for measurement o f fatty acid content by GLC and/or phosphorus determination.

III. Results and discussion A. Pressure~rea curves o f pure filrn, s

Pressure-area curves o f the individual PCs show that the films fall into three types,

212

R.I¢. Evans, J. Tinoco, Sterols and phosphatidylcholines in monolayers

according to molecular area (fig. 1). The saturated 16 : 0 - 2 0 : 0 PC occupied the smallest area/molecule and could withstand much higher surface pressures than any of the other PC examined. At room temperature, small molecular areas are typical of saturated PC o f chain length longer than 16 carbons, such as 18 : 0 - 1 8 : 0 PC [ 1 , 3 3 37] and 22 : 0 - 2 2 : 0 PC [34]. At room temperature, saturated PC o f chain length shorter than 16 carbons form expanded films with large molecular areas [35]. PC with one or two double bonds (16 : 0 - 2 0 : ln9 and 16 : 0 - 2 0 : 2n6 PC) produced curves very similar to each other, with molecular areas much greater than those of 16 : 0 - 2 0 : 0 PC. The addition o f one more double bond produced a further large increase in area (16 : 0 - 2 0 : 3n3, 16 : 0 - 2 0 : 3n6 and 16 : 0 - 2 0 : 3n9 PC). A similar phenomenon was reported earlier in monoalyers o f 16 : 0 - 1 8 : ln9 PC, 16 : 0 - 1 8 : 2n6 PC and 16 16 : 0 - 1 8 : 3n3 PC, in which molecular areas of the mono- and diunsaturated PC were similar and much smaller than that of 16 : 0 - 1 8 : 3n3 PC [36]. In all these examples, the double bonds have been methylene-interrupted, as is common in animal lipids. It is not known whether similar molecular areas would also be obtained if the double bonds were distributed differently within the side chain. Molecules with 4 or 5 double bonds (16 : 0 - 2 0 : 4n6 PC and 16 : 0 - 2 0 : 5n3 PC)

70 I

16:0-20:0 PC

60 16:0- 20:In9PC

t | /

l I CHOLESTEROL--II,

16:0-20:2n6 PC

~ ~ V.

• \

16:0- 20:3n9PC

DESMOSTEROL

:0-20:3n5 PC

I0

0

PC

20

40

60 SO • I00 AREA/MOLECULE, ~2

120

140

160

Fig. 1. Pressure-area curves (,f sterols and PC at the air-water interface at 22 _+ 2°C. The subDhase

was glass-distilled water, pH about 5.1.

R.W. Evans, J. l~noco, Sterols and phosphatidylcholines in monolayers

213

had molecular areas in the same range as those with three double bonds, that is, additional double bonds did not increase molecular area. The 16 : 0 - 2 0 : 3n9 PC produced the largest area of all, but it should be noted that about 6% of other components were present (see Experimental). Essential fatty acid deficiency in animals is characterized by the presence of 20 : 3n9 in tissue phospholipids. In phosphatidylcholine liposomes prepared from essential fatty acid-deficient rat livers, Moore et al. [38] reported greater Na+ permeability than in liposomes from normal rat liver phosphatidylcholines. These systems involved very complex mixtures of lipids so that it is not possible to attribute increased permeability to 20 : 3n9 PC alone. Yet the apparent large molecular area of 16 : 0 - 2 0 : 3n9 PC in monolayers does indicate that the molecules are relatively inefficiently packed. B. Molecular areas in m i x e d monolayers

The areas/molecule of most of the mixed monolayers were smaller than the sum of the areas of the pure components. This deficit or shrinkage is termed condensation and table 3 shows values of condensations measured at 30 and 10 dyn/cm. Cholesterol and desmosterol condensed the phosphatidylcholines to a similar extent. Monolayers of all unsaturated phosphatidylcholines condensed most at sterol concentrations between 40 and 60 mol%, but 16 : 0 - 2 0 : 0 PC condensed most in mixtures containing 2 0 - 3 0 mol% of either sterol. Fig. 2 presents the areas of mixed monolayers plotted versus composition for 16 : 0 - 2 0 : 0 PC and 16 : 0 - 2 0 : 4n6 PC with each sterol. Condensation of 16 : 0 - 2 0 4n6 PC with either sterol was greatest at the middle of the concentration range, but that of 16 : 0 - 2 0 : 0 PC was greatest at the higher PC concentrations. To emphasize the influence of concentration, condensations were plotted versus sterol concentrations (fig. 3). Fig. 3A shows that the interactions of cholesterol and desmosterol with 16 : 0 - 2 0 : 0 PC were qualitatively similar in that condensations of both were greatest at about 70 mol% PC. The unsaturated PC all condensed to the greatest extent at 4 0 - 6 0 % PC (fig. 3B,C). Earlier work had also shown that cholesterol condensed most with unsaturated PC at 4 0 - 6 0 mol % PC (Fig. 3D). In addition, the behaviour of the saturated PC (16:0-16:0PC, 18:0-18:0PC [36], 1 6 : 0 - 2 0 : 0 P C [31]) mimicked PC 1 6 : 0 20 : 0 in exhibiting maximum condensation at about 70 mol% PC (fig. 3D). Although all the examples given here of PC that interact most with sterols at concentrations of about 70 mol/PC are saturated, the physical state of the PC molecule and not its saturation may be a regulating factor in sterol-PC interaction. The nature of its side chains, including degree of unsaturation, influences the physical state of a PC molecule and for the examples cited, at room temperature, the physical state of the saturated molecules probably differs from that of the unsaturated molecules. The results of Demel et al. [33] indicate that, at room temperature, the short chain saturated 14 : 0 - 1 4 : 0 PC, which forms an expanded film at this temperature, interacts most with cholesterol when the two components are present in approximately equimolar amounts. Thus expanded PC films, either saturated or unsaturated, condense most when the molar ratio of PC to sterol is about 1 : 1, but compact films of saturated PC condense most at a molar ratio of about 2 PC to 1 sterol.

Cholesterol Desmosterol Cholesterol Desmosterol Cholesterol Desmosterol Cholesterol Desmmterol Cholesterol Desmosterol Cholesterol Desmostetol Cholesterol Desmosterol Cholesterol Demosterol Cholesterol Desmosterol Cholesterol Desmosterol Cholesterol Desmosterol Cholesterol Desmosteroi Cholesterol Desmosterol Cholesterol Desmosterol Cholesterol Desmosterol Cholesterol Desmosterol

PC16 : 0 - 2 0 : 0

10

30

o

Dynes/on

113.6

180.0

119.6

107.0

106.2

85.8

83.8

64.9

81.1

80.2

87.6

78.8

79.0

65.9

62.9

52.7

0

Total area, pure PC

9.6 3.4 2.7 1.0 3.2 3.6 3.6 4.9 0.4 1.4 4.5 4.4 5.8 6.1 2A 1.1 9.6 3.2 3.2 2.2 5.5 5.5 5.2 8.0 0.0 1.2 4.9 6.0 7.1 7.9 0.0 -0.3

0.2

Condensation

Mol fraction of sterol

10.8 5.1 3.0 2.0 6.5 5.7 6.6 7.0 2.0 2.8 5.2 6.5 7.1 7.2 3.7 3.7 13.6 6.5 4.6 3.3 9.9 8.0 9.2 10.8 1.8 4.2 6.1 7.6 8.3 9.1 1.2 3.4

0.3 6.7 4.9 4.7 4.7 8.0 6.4 6.6 7.3 2.7 3.9 7.1 8.9 8.6 9.8 6.0 5.5 8.7 6.7 7.2 6.4 12.3 7.9 9.9 11.9 2.2 5.5 6.7 10.4 10.7 12.2 1.9 4.7

0.4 3.9 3.3 7.0 5.4 7.3 7.1 8.3 8.4 3.4 5.5 8.6 8.5 9.3 8.7 6.9 5.6 5.8 5.5 11.2 8.6 11.4 10.6 12.4 13.0 4.7 6.7 9.0 9.9 10.9 10.8 5.8 4.4

0.5

aideal area minus observed areas are given in A 2/molecule. Mole fractions varied within ± 10% of stated values.

PC16 : 0 - 2 0 : 5n3

PCI6 : 0 - 2 0 : 4n6

PC16 : 0 - 2 0 : 3n9

1)(216 : 0 - 2 0 : 3n6

PC16 : 0 - 2 0 : 3n3

PC16 : 0 - 2 0 : 2n6

PCI6 : 0 - 2 0 : In9

PC16 : 0 - 2 0 : 0

PC16 : 0 - 2 0 : 5n3

PC16 : 0 - 2 0 : 4n6

PCI6 : 0 - 2 0 : 3n9

PC16 : 0 - 2 0 : 3n6

PC16 : 0 - 2 0 : 3n3

PCI6 : 0 - 2 0 : 2n6

PC16 : 0 - 2 0 : ln9

Sterol

in mixed monolayers containing sterols and phosphatidylcholines a.

PC

Decreases in area/molecule, ~ 2 ,

Table 3

3.1 2.3 6.5 6.0 8.4 5.5 7.0 8.4 4.2 5.0 7.6 8.2 6.8 8.3 6.7 5.8 3.5 4.0 11.0 8.0 11.9 10.2 11.6 13.7 5.4 7.5 8.1 11.4 10.0 11.5 6.7 6.5

0.6 1.0

39.4 37.5 39.4 37.5 39.4 37.5 39.4 37.5 "39.4 37.5 39.4 37.5 39.4 37.5 39.4 37.5 42.5 41.5 42.5 41.5 42.5 41.5 42.5 41.5 42.5 41.5 42.5 41.5 42.5 41.5 42.5 41.5

0.8

0.9 -0.4 3.9 4.5 3.7 2.4 3.4 4.8 2.5 4.6 5.6 4.0 3.4 5.1 3.9 3.2 0.7 0.1 7.2 8.8 6.4 6.2 6.4 9.2 3.7 7.8 6.5 6.6 6.1 8.8 3.4 3.8

Total area, pure sterol r~

to

R.W. Evans, J. l)'noco, Sterols and phosphatidylcholines in monolayers 60

A

215

B

50 40 30 20 10 0 tff ._1

50

5O

C H O L E S T E R O L 16 : 0--20 : 0 PC

D E S M O S T E R O L 16 : 0 - 2 0 : 0 PC

LU

._1 0

9° r c

D

80 tu n"

70 60 50 40 30 20 10 0

50

CHOLESTEROL

50 16 : 0--20 : 4n6 PC

DESMOSTEROL

16 : 0 20 : 4 n6 PC

C O N C E N T R A T I O N , MOL%

Fig. 2. Variation of the mean molecular area as a function of composition in mixed monoiayers of PCs and sterols. Surface pressure, 30 dyn/cm. The subphase was glass-distilledwater, pH about 5.1, temperature 22 -+2°C. Diagonal lines represent contribution of PC or sterol to ideal area of mixed monolayer.

C Effects o f surface pressure on extent o f condensation in monolayers containingPC and cholesterol Fig. 4 shows the variation of condensation with surface pressure in equimolar mixtures of cholesterol and PC. In mixtures of other mole ratios, curves o f the same shape were found in nearly every case. In mixtures o f PC with cholesterol, the surface pressure at which m a x i m u m condensation occurred depended upon the positions of.the double bonds. Those PC that possessed a long section of saturated chain near the carboxyl group showed maximum con-

I

80

~

B

i

802

100

-2 L

0

!

12

A

I

60

i

40

I

40

MOt% CHOLESTEROL

60

I

MOL% STEROL

, "~

,

20

I

20

" ~

L

0 !~'~

'

0 1

\~

8 0 / ~'-'r-~

C

40 MOt% CHOLESTEROL

60

I

20

0

Fig. 3. Variation of condensation as a function of composition in mixed monolayers of sterols and PCs. Temperature, 22 ± 2°C. Subphase was glass-distilled water, pH about 5.1. Surface pressure, 30 dyn/cm in parts A, B, C; 40 d y n / c m in part D. Data in part D were taken from the literature [31,36]. Vertical bars in A represent S.E.M. of two or three determinations. S.E.M. in B and C never exceeded 0.6 A 2 . Values of D not available. A: (e) 16 : 0 20 : 0 PC-cholesterol; (o) 16 : 0 - 2 0 : 0 PC-desmosterol. B: (A) 16 : 0 - 2 0 : 3n9 PC; (o) 16 : 0 - 2 0 : 3n3 PC; (z~) 16 : 0 - 2 0 : l n 9 PC; (o) 16 : 0 - 2 0 : 5n3 PC. C: (e) 16 : 0 - 2 0 : 4n6 PC; (zx) 16 : 0 - 2 0 : 2n6 PC; (o) 16 : 0 - 2 0 : 3n6 PC. D: (~) 18 : 0 - 1 8 : 0 PC; (o) 16 : 0 - 1 6 : 0 PC; (zx) 16 : 0 - 2 0 : 0 PC; (m) 16 : 0 - 1 8 : 2n6 PC, (e) 16 : 0 - 1 8 : I n 9 PC; (A) 16 : 0 - 2 0 : 4n6 PC.

~)"

~

100 -2 L

o

lO

12

R.W. Evans, J. Tinoco, Sterols and phosphatidylcholines in monolayers 16

217

B

A

14 12

lO

4

~ 2 / I

I

I

;

10

20

30

40

Z 016 F-

~} 1

I

I

I

20

30

40

D

z z 8

lo

i

i

10

20

i

30

,i 40

I

I

I

I

10

20

30

40

SURFACE PRESSURE, DYNES/CM Fig. 4. Variation of condensation as a function of surface pressure at 22 -+ 2°C for equimolar mixtures of cholesterol and PCs. Subphase was glass-distilled water, pH about 5.1. Vertical bars represent S.E.M. of two or three determinations. A: (o) 16 : 0 - 2 0 : ln9 PC; (e) 16 : 0 - 2 0 : 0 PC. B: (~) 16 : 0 - 2 0 : 2n6 PC; (A) 16 : 0 - 2 0 : 3n3 PC;C: (=) 16 : 0 - 2 0 : 3n9 PC;(o) 16 : 0 - 2 0 : 3n6 PC; D: (~7)16 : 0 - 2 0 : 4n6 PC; (v) 16 : 0 - 2 0 : 5n3 PC. densations at pressures o f 5 d y n / c m or less (Fig. 4A,B). This group included molecules which c o n t a i n e d 20 : 0, 20 : l n 9 and 20 : 2 n 6 , all o f which have 10 carbons or m o r e o f unsaturated chain length adjacent to the carboxyl group. In molecules w i t h double bonds closer to the c a r b o x y l group, m a x i m u m condensations o c c u r r e d at surface pressures o f 15 or 20 d y n / c m (Fig. 4 C ~ ) ) . In n o n e o f the PC did m a x i m u m condensation occur above 20 d y n / c m .

D. Effects o f surface pressure on extent o f condensations in monolayers containing PC and desmosterol The behaviour o f desmosterol was similar to, but n o t identical withe that o f choles-

218

R.W. Evans, J. l~noco, Sterols and phosphatidylcholines in monolayers

terol in mixed monolayers (data not shown). Condensation of 16 : 0 - 2 0 : 0 PC was only slightly affected by surface pressure, at all mole ratios. For 16 : 0 - 2 0 : ln9 PC mixtures, maximum condensations were observed at 5 to 10 dyn/cm, pressures slightly higher than those that produced maximum condensations with cholesterol. The resuits for 16 : 0 - 2 0 : 2n6 PC and 16 : 0 - 2 0 : 3n3 PC closely resembled those obtained with cholesterol: maximum condensation occurring at 5 dyn/cm or less. Maximum condensation in mixtures with 16 : 0 - 2 0 : 3n6 or 16 : 0 - 2 0 : 3n9 occurred from 5 20 dyn/cm, depending on the mole ratios of the components. In mixtures of 16 : 0 20 : 4n6 with desmosterol, maximum condensation occurred at 10 dyn/cm. Mixtures of 16 : 0 - 2 0 : 5n3 PC with desmosterol exhibited maximum condensation at 20 dyn/ cm, as was observed for 16 : 0 - 2 0 : 5n3 and cholesterol. Cholesterol and desmosterol resembled each other in that maximum condensations never occurred at pressures above 20 dyn/cm. At present it is not clear how surface pressure in monolayers relates to the condition oflipids in artificial bilayers or in natural membranes. Demel et al. [39] calibrated the pressure-dependence of several phospholipases by measuring their activities upon monolayers of pure synthetic phospholipids at different surface pressures. Only those enzymes active upon monolayers at pressures above 31 dyn/cm were also capable ofhydrolyzing the same lipids in erythrocytes. This indicated that the lipids of the erythrocyte were as accessible as the same lipids in monolayers subjected to pressures above 31 dyn/cm. However~ Cornwell and Patil [2] have estimated, by completely different reasoning, that the physiological surface pressure lies in the range 10-25 dyn/cm. It seems possible that the conformations of lipids in membranes may correspond to a wide range of surface pressures in monolayers. It is clear from the data in fig. 4 that the condensations of synthetic PCs with cholesterol are greatly influenced by surface pressure and by the positions of double bonds in the PC. Mixtures of PCs with desmosterol were similar to, but distinguishable from, their mixtures with cholesterol. These differences must result from the differences in the side chains of the sterols. Although the function of the sterol side chain in membranes is not understood, the cholesterol side chain is preferred to that of desmosterol in most animal tissues. At least the final step in cholesterol biosynthesis evidently has survived much evolutionary selection pressure. We had expected that there would be a greater difference between the interactions of cholesterol and desmosterol with polyunsaturated PC, because desmosterol has been found in tissues unusually rich in polyunsaturated PC. But the condensations of desmosterol with 16 : 0 - 2 0 : 4n6 PC or 16 : 0 - 2 0 : 5n3 PC were not remarkably different from those obtained with cholesterol. Therefore, the association of desmosterol with highly unsaturated lipids in brain and spermatozoa does not seem to be a result of a special interaction between desmosterol and polyunsaturated 20-carbon PC. However, brain and spermatozoa also contain phospholipids very rich in 22 : 6n3, and these may behave differently with desmosterol than with cholesterol. The behaviour of 16 : 0 - 2 0 : 3n9 PC with cholesterol was similar to that of either 16 : 0 - 2 0 : 3n6 PC or 16 : 0 - 2 0 : 4n6 PC. Therefore, the peculiar permeability prop-

R. I¢. Evans, J. 1~noco, Sterols and phosphatidylcholines in monolayers

219

erties o f membranes during essential fatty acid deficiency probably cannot be attributed to differences between the interactions of normal and essential fatty acid-deficient PC with sterols. However, both the pure film of 16 : 0 - 2 0 : 3n9 PC and its mixtures with sterols had unusually large molecular areas. PCs containing a 20 : 3n9 chain have been examined before in monolayers, but in both cases they were included in very complex mixtures of many molecular species of PC obtained from essential fatty acid-deficient rats [32,40]. In both cases, the mixtures had molecular areas smaller than those of PC from normal rats. This contrast with the large area of our relatively pure 16 : 0 - 2 0 : 3 n 9 PC may be either an experimental artifact or it may be real. It is possible that in complex mixtures, different PC may condense with each other, or that the other constituents of these mixtures were responsible for the small average molecular areas of the mixtures. It is noteworthy that the naturally most abundant PC molecules (those containing 20 : 4n6, 20 : 5n3, 20 : 3n9 and 20 : 3n6) exhibited the largest molecular areas in pure films. Furthermore, these molecules also condensed most with sterols at pressures of 15 and 20 dyn/cm, in contrast to the behaviour of the other PC which condensed most at lower surface pressures. Possibly, molecules which form expanded monomolecular films and condense maximally with sterols at relatively high surface pressures are the most suitable for membrane formation.

Acknowledgements This work was supported in part by USPHS Grants AM 12024 and AM 10166. We are grateful to Ms. Ruth Babcock, Mr. Peter Miljanich and Ms. Barbara Medwadowski for the purifications oficosatrienoic acid (20 : 3n9) and icosapentaenoic acid (20 : 5n3). We thank Ms. E.A.D. Evans, Ms. Lynn Conway and Ms. Pearl Kadota for preparation of the manuscript. We greatly appreciate the advice and encouragement of Dr. M.A. Williams, in whose laboratory this work was performed.

References [1] R.A.Demel, J. Am. Oil Chem. Soc. 45 (1968) 305 [2] D.G. Cornwell and G.S. Patti, in Polyunsaturated Fatty Acids, W.4-I. Kunau and R.T. Holman (eds.) Amer. Oil Chem. Soc. (1977) 105 [3] G. Weitzel, A.-M. Fretzdorff and S. Heller, Hoppe-Seyler'sZ. Physiol. Chem. 288 (1951) 189 [4] G. Weitzel, A.-M. Fretzdorffand S. Heller, Hoppe-Seyler'sZ. Physiol. Chem. 288 (1951) 200 [5] P.G. Barton and F.D. Gunstone, J. Biol. Chem. 250 (1975) 4470 [6] R.F. Addison, R.G. Aekrnan and J. Hingley, J. Fish. Res. Bd. Canada 25 (1968) 2083 [7] E.M. Kreps, M.A. Chebotaf~va and V.N. Akulin, Comp. Biochem. Physiol. 31 (1969) 419 [8] R.F. Addison and R.G. Ackman, Can. J. Biochem. 49 (1971) 873 [9] R.L.Lyman, G. Sheehan and J. Tinoc0, Can. J. Biochem. 49 (1971) 71

220 [10] [11] [12] [13] [ 14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29]

[30] [31] [32] [33] [34] |35] [36] [37] [38] [39] [40]

R.W. Evans, J. Tinoco, Sterols and phosphatidylcholines in monolayers D.A. White, in, Form and Function of Phospholipids, G.B. Ansel], R.M.C. Dawson and J.N. Hawthorne (eds.), Elsevier (1973) 468 G.E. Scheier and M.A. Williams, Fed. Proc. 21 (1962) 393 R.R. Brenner and A.M. Nervi, J. Lipid Res. 6 (1965) 363 R.T. Holman, in Progress in the Chemistry of Fats and Other Lipids, Vol. IX, R.T. Holman (Ed.), Pergamon, (1968) 275 G. Bleau and W.J.A. VandanHeuvel, Steroids 24 (1974) 549 R. FumagaUi, E. Grossi, P. Paoletti and R. Paoletti, J. Neurochem. 11 (1964) 561 R.G. Dennick, K.J. Worthington, D.R. Abramovich and P.D.G. Dean, J. Nuerochem. 22 (1974) 1019 E. Kuhn, W. Dorow, W. Kahlke and H. Pfisterer, Klin. Wschr. 46 0968) 1043 D. Sefler and E. Kuhn, Z. Klin. Chem. Klin. Biochem. 9 (1971) 245 W. Fiehn, D. Seiler, E. Kuhn and D. Bartels, Eur. J. Clin. Invest. 5 (1975) 327 L. Svennerholm, J. Lipid Res. 9 (1968) 570 A. Poulos, A. Darin-Bennett and I.G. White, Comp. Biochem. Physiol. 46B (1973) 541 J. Tinoco, S.M. Hopkins, D.J. Mclntosh, G. Sheehan and R.L. Lyman, Lipids 2 (1967) 479 P.S. Cben, Jr., T.Y. Tofibara and H. Warner, Anal. Chem. 28 (1956) 1756 D.R. Snider and E.D. Clegg, J. Anim. Sic. 40 (1975) 269 V.P. Skipski, R.F. Peterson and M. Barclay, Biochem. J. 90 (1964) 374 C. Long and I.F. Penny, Biochem. J. 65 (1957) 382 Z. Selinger and Y. Lapidot, J. Lipid Res. 7 (1966) 174 E. Cubero Robles and D. van den Berg, Biochim. Biophys. Acta 187 (1969) 520 IUPAC-IUB Commission on Biochemical Nomenclature (CBN members: O. HoffmannOstenhof, W.E. Cohn, A.E. Braunstein, H.B.F. Dixon, B.L. Horecker, W.B. Jakoby, P. Karlson, W. Klyne, C. Liebecq and E.C. Webb), Lipids 12 (1977) 455 R.A. Demel, W.S.M. Geurts van Kessel and L.L.M. van Deenen, Biochim. Biophys, Acta 266 (1972) 26 D. Ghosh, M.A. Williams and J. Tinoco, Biochim. Biophys. Acta 291 (1973) 351 J. Tinoco and D.J. Mclntosh, Chem. Phys. Lipids 4 (1970) 72 R.A. Demel, LL.M. van Deenen and B.A. Pethica, Biochim. Biophys. Acta 135 (1967) 11 M.C. Phillips and D. Chapman, Biochim. Biophys. Acta 163 (1968) 301 P. Joos and R.A. Demel, Biochim. Biophys. Acta 183 (1969) 447 D. Ghosh and J. Tinoco, Biochim. Biophys. Acta 266 (1972) 41 M.C. Phillips, H. Hauser and F. Paltauf, Chem. Phys. Lipids 8 (1972) 127 J.L. Moore, T. Richardson and H.F. DeLuca, Chem. Phys. Lipids 3 (1969) 39 R.A. Demel, W.S.M. Geurts van Kessel, R.F.A. Zwaal, B. Roelofsen and L.L.M. van Deenen, Biochim. Biophys. Acta 406 (1975) 97 L.M.G.van Golde and LL.M. van Deenen, Biochim. Biophys. Acta 125 (1966) 496