Structural alterations of phospholipid film domain morphology induced by cholesterol

Transcription

1 ndian Journal of Biochemistry & Biophysics Vol. 40, April 2003, pp Structural alterations of phospholipid film domain morphology induced by cholesterol A K Panda l,2*, A Hume 2, K Nag 2,3, R R Harbottle 2 and N 0 Petersen= 'Department of Chemistry, Behala College, Kolkata , ndia 2Department of Chemistry, The University of Western Ontario, London, ON-N6A 5B7,Canada 3Dept. of Biochemistry, Memorial University of Newfoundland, St. John's, Newfoundland AB 3X9, Canada Received 23 July 2002; revised 31 January 2003 Structures of the monolayer films of dipalmitoylphosphatidylcholine (DPPC) mixed with different amounts of cholesterol were studied at air-water interface using surface pressure-area measurements, epifluorescence microscopy and atomic force microscopy (AFM). Pure DPPC, cholesterol or DPPC-cholesterol mixtures were dissolved in organic solvents with a small amount of fluorescently labeled phospholipid probe (NBD-PC) and spread onto the air-water interface. Surface pressure-area isotherms and epifluorescence microscopy of such films at the air-water interface suggested that DPPC undergoes a gas to fluid to condensed phase transition, while cholesterol undergoes a gas to solid-like transition. A shift of the surface pressure-area curve to lower area per molecule was observed when cholesterol was mixed with DPPC. Epifluorescence microscopy showed the formation of spiral shaped domains for mixed monolayers. ncrease in cholesterol content abolished domain characteristics possibly due to fluidizing property of cholesterol. AFM measurements of monolayers, transferred onto freshly cleaved mica by Langmuir-Blodgett technique, revealed the alterations caused by cholesterol on the gel and fluid domains of such films. AFM measurements re-established similar trend in domain characteristics as evidenced in epifluorescence microscopy. Gas exchange in a lung occurs in the alveoli, which provide a large, wet surface area exposed to air. The alveoli are nearly spherical with such a high curvature that they should collapse or fill with liquid. Fortunately, in healthy lungs a thin insoluble film, known as pulmonary surfactant, lines the surface of the alveoli and prevents their collapse. The reduction of surface tension at alveolar surface permits lung expansion on aspiration and prevents lung collapse on expiratiorr'". The most abundant phospholipid in pulmonary surfactant is dipalmitoylphosphatidylcholine (OPPC) which comprises -35% of mammalian surfactant by weights. Cholesterol is another.component which makes up approximately 8% (by wt.) of surfactant". Cholesterol is also an important constituent of biological membranes and is known to control fluidity of the lipid bilayer of the cell membrane. At low concentrations (up to 5%, by wt.) in the bilayer, it reduces the enthalpy change and broadens the host lipid phase transition by making the membrane more fluid in the gel phase and less fluid in the liquid *Corresponding author AKP:akpanclal@yailoo.com NOP: petersen@lwo.ca crystalline phase". At higher concentrations (above 10% by wt.), cholesterol becomes insoluble and phase separates 10. Although monolayers are not necessarily a good model for biological membranes, the properties of cholesterol in a mixed monolayer may provide additional important information like phase transition, mutual miscibility, energetics of mixing, etc. Cholesterol influences lateral domain formation, thus acts as a "lineactant" (that reduces line tension) in the boundary region between two-dimensional gel and liquid crystalline phases". Lipids in biological membranes are generally in a liquid crystalline state, which corresponds to an extended state of the monolayer, although there are a few exceptions'<". Monolayer mixtures of cholesterol and phospholipids have been used as models for lipid distributions in biological membranes Also, epifluorescence microscopic studies have allowed direct observation of the various phases that exist in the mixed monolayers including mixtures of DPPC and cholesterol at the air-water interface';". However, a detailed study of such systems using epifluorescence and atomic force microscopy has not been undertaken, Moreover, a comparative study between two techniques has also not been reported to the best of our knowledge,

2 PANDA et al.: MORPHOLOGY OF MXED DPPC-CHOLESTEROL FLMS 115 nthis paper, we report a combined epifluorescence andafm study on the monolayer with mixtures of DPPC and cholesterol spread on the air-water interface.the mixed monolayers were compressed slowlyand phase transition from liquid expanded (LE) to liquid condensed (LC) state was monitored throughthe formation of probe-excluded domains (appearedblack, when seen under epifluorescence microscopy). Compressibilities of the films at differentcompositions were also measured. Surface films were deposited by Langmuir-Blodgett techniques onto freshly cleaved mica. Surface morphologyof this the transferred monolayer were monitoredby AFM. Materialsand Methods Materials 1,2-Dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) andthe fluorescent lipid probe -palmitoyl-2-{ 12-[(7- nitrobenz-2-oxa-l,3-diazo-4- y) amino] dodecanoy l}- sn-glycero-3-phosphocholine (NBD-PC) were obtained from Avanti Polar Lipids nc.,usa. Cholesten-3-01(cholesterol) was obtained from Sigma ChemicalCo., USA. All the chemicals were 99% pure andused as received. Stocksolutions of each compound at 1.0 mg ml- ' in chloroform:methanol (3: 1, vv) were used to prepare mixtures of DPPC:cholesterol:NBD-PC at 98:1:1,96:3:1, 93:6:1, and 89:10:1, w/w. These were storedat -20 e. Double distilled water was used as the subphase. Thesecond distillation was done using dilute alkaline " permanganatesolution to oxidize the organic traces. All the compressions and deposits as well as fluorescence and atomic force microscopy experimentswere done at 25 e. Surface pressure(n)-area(a) isotherms A Langmuir trough manufactured by Kibron (JLTroughSX, Finland) and having the total surface area of 125 crrr' was used throughout. Surface pressurewas detected using a Wilhelmy balance wire probe.monolayers were spread onto 90 ml of doubly distil1~dwater (ddh 2 0) subphase using a 10 f.ll Hamiltonsyringe and a wait period of 30 min. was introduced to allow for solvent evaporation and equilibration of the monolayers on the surface. They werethen compressed at a rate of 0.04 nm 2 molecule"! min" and the pressure-area isotherms were recorded. Detailshave been described elsewhere '6. Epifluorescence microscopy Monolayers were viewed with a 35x objective lens with a numerical aperture of 0.4 and a l x ocular lens of a Zeiss epifluorescence microscope equipped with a 100 W mercury lamp. mages of the monolayers were recorded with the Retiga" digital camera attached to a Stardancer 2 intensifier and analyzed using the Northern Eclipse (V) imaging software. Transfer of Langmuir Blodgett films onto mica Films of DPPC: cholesterol (98:1, 96:3, 93:6, 89: 10) containing 1 wt % NBD-PC were compressed to pressures at 7 and 10 mnm- '. Once the desired pressure was obtained, a waiting period of 10 min was introduced before the freshly cleaved mica was raised vertically through the monolayers and out of the trough at a rate of 10 mm min -. Compressing the film at equal rate compensated the transfer of material onto mica. All deposited monolayers were then visualized using contact mode atomic force microscopy in air within 2 hr of deposition S. Atomic force microscopy (AFM) All Langmuir-Blodgett films were analyzed using contact mode AFM. mages were obtained using a Nanoscope a equipped with a multimode head and an oxide sharpened silicon nitride cantilever with a length of 200 urn and a spring constant of 0.06 Nm '. The films were scanned at rates ranging from 0.1 to 2 Hz with the higher scan rates used to scan smaller scan sizes 19. Results and Discussion Surface pressure-area isotherms Pressure-area isotherms of pure DPPC, cholesterol, and mixed molar proportions of DPPC:cholesterol (99:1, 97:3, 94:6, 90:10, w/w) were constructed in order to characterize the effects of cholesterol on DPPC monolayers. Fig. 1 shows the isotherms of pure DPPC, cholesterol and their 90: 10 mixtures as representative plots. Some of the quantitative features of pure components as well as their mixtures are summarized in Table 1. The pure DPPC isotherm showed the typical first order phase transition from liquid expanded (LE) to liquid condensed (LC) represented by the plateau region at rt = 6 mnm '. n contrast, the pure cholesterol isotherm showed a sharp increase of surface pressure at approx nm 2 molecule", representing a phase transition from an expanded gas phase (G) to a solid condensed phase (SC), which was in agreement with earlier results 9,'4.'.5, Mixed monolayers behaved differently.

3 116 NDAN J. BOCHEM. BOPHYS., VOL. 40, APRL ~ , 55 "';" 45 E z.s 35 OJ :; tn ~ 25 a. OJ ~ 15 't: :::> (/J "' -, --_/,, _5+- L- ~L_ ~ ~ ~ Area (nm 2 ) per molecule Fig. -Surface pressure-area isotherms of pure DPPC (-), pure cholesterol (---), and mixed molar proportions of DPPC:cholesterol 90:1O,w/w ( ) [All films were compressed at a rate of 0.04 nm" moleculelmin ": temperature 25 C] Table --Quantitative analysis of the pure DPPC isotherm and the DPPC:cholesterol isotherms (99:1), (97:3), (94:6), and (90: 10) w/w. sotherm KT/m(mNrl ~A/nm2 T at midpoint 1.0 LC LE rnolecule " of plateau/ mnm- 1 DPPC DPPC:chol (99: 1) DPPC:chol (97:3) DPPC:chol (94:6) DPPC:chol (90: 10) The surface pressure-area curves were in between two pure components. A gradual shift of the curves towards lower surface area was observed when the cholesterol content was increased. As increasing amounts of cholesterol were added to DPPC monolayer, the density of two phases became more alike resulting in a decrease in length of phase transition. Similarly, as increased amounts of cholesterol were present in DPPC monolayer, phase transition from LE to LC phase occurred over a larger range of surface pressure indicating that cholesterol decreased the co-operativity of this transition. This effect of cholesterol on DPPC isotherms was consistent with that reported by Worthman et al', A quantitative analysis was performed on the isotherms measured in Fig. 1. Slope measurements of LE and LC regions were suitably processed, according to equation 1, which shows the formula used to calculate the compressibility, KT, of LE and LC regions.".... (1) where A is the area of the monolayer per molecule and 7[ is the surface pressure. Table 1 summarizes the quantitative analyses of the isotherms, which also illustrates that addition of cholesterol from 1-10% did not measurably alter the compressibility, KT, of LE or LC region of a DPPC:cholesterol monolayer. Similarly, cholesterol showed no effect on the midpoint pressure of the phase transition. Finally, as increasing amounts of cholesterol were added, the length of the phase transition from LE to LC of DPPC decreased and eventually was not measurable with 10% cholesterol. The observation was consistent with the effect of cholesterol on the thermal phase transition of DPPC 9,14. As cholesterol condensed LE phase and fluidized LC phase in DPPC mono!ayers, an increase in the compressibility of LC region and a corresponding decrease in the compressibility of LE region took place. Decrease in the cooperativity of LE-LC phase transition resulted with the increase in cholesterol content, for which the midpoint pressure remained unchanged. Fluorescence image analysis Monolayer of DPPC:cholestero! (98: 1) contammg 1 wt% NBD-PC was compressed at a rate of 0.04 nm' molecule." min- 1 and was visualized using epifluorescence microscope at surface pressures from 7 to 11 mnm- 1 (Fig. 2). The fluorescent images in Fig 2 show a variety of LC domains represented by the darker regions. Fig. 2 B, C' and D indicate that LC domains are aggregating and forming different complex shapes. Thus, from Fig. 2, it appears that the amount of LC regions did not decrease as the pressure increased; they occupied different spaces and formed different shapes. The spiral shapes of LC domains, resolved by the fluorescence image, were not seen in the AFM images (to be shown later). These structures

4 PANDA et al.: MORPHOLOGY OF MXED DPPC-CHOLESTEROL FLMS S '" 20 '"S 0 ~ ;..! 10 S ::: z, nterval Area of Domains (Jln) Fig. 2-Fluorescence images of a monolayer of DPPC: cholesterol(98:, w/w) containing wt% NBD-PC [At pressures of (A):?, (B):8.5, (C): 10, and (D):llmNm- l ] Fig.4-Fluorescence images of a monolayer of DPPC:cholesterol 96:3(A,D), 93:6 (B, E) and 89: 10 (C, F) containing wt% NBD- PC. [magesa, Band C were taken at surface pressures of 7 mn m", whileimages D, E and Fat 10 mnm-l] Fig. 3--Histogram plot showing the number of LC domains with a given area measured in a monolayer of DPPC:cholesterol (98:1) containing 1 mol % NBD-PC at a surface pressure of 7mNm- 1 were also observed and analyzed in details previously by Gaub et al. 21. The areas of LC domains in the fluorescent image were measured using Northern Eclipse imaging software and a histogram plot of the areas was constructed (Fig. 3). The average size of LC domains in fluorescence image was measured to be approx. 29 /J.m2. The histogram plot (Fig. 3) of fluorescence image demonstrated the wide range of sizes of LC domains in the DPPC:cholesterol monolayer. The largest number of domains has an area between 25 " and 35 /J.m2.The shape of histogram is representative of a unimodal distribution suggesting that cholesterol did not induce the formation of any separate domains. Fig. 4 shows the fluorescence images for the mixed monolayers containing DPPC:cholesterol:NBD-PC at different wt. ratios (96:3:1, 93:6:1 and 89:10:1) taken at two different surface pressures [7mNm- 1 (Fig. 4 A, B and C) and 10.0 mnm- 1 (Fig. 4 0, E and F)]. The images in samples with 3 wt% cholesterol (Fig. 4 A and D) showed circular black spots, corresponding to LC domains which did not appear to be in direct contact with one another at both pressures. The image at 10 mnm- 1 (Fig. 40) appeared to have a larger number of black spots, and thus a larger number of LC domains, compafed to the lower pressure image (Fig. 4A). Thus, as the pressure was increased from 7 ml-lm' to 10 ml-lm", there appeared to be an increase in the number LC regions. The fluorescence images for mixed mono layers containing 6% cholesterol showed only brighter areas with no dark areas

5 118 NDAN J. BOCHEM. BOPHYS., VOL. 40, APRL 2003 (Fig. 4 B and E). From the fluorescence images for mixed monolayer containing 10% cholesterol (Fig. 4 C and F), it appeared that at both pressures, a third single phase was formed. One possible explanation for the decrease in black areas in the fluorescent images is that cholesterol retarded the formation of LC phases in DPPC monolayers. Another possible explanation is that the fluorescent probe was unable to distinguish between two phases as the amount of cholesterol was increased. The fluorescent probe might have been mixed evenly throughout the monolayers and not partitioned between the two phases. Without further analysis of the monolayer, the fluorescent images were unable to discern between two possible explanations. Nevertheless, decreasing difference in the density between two phases supported the possibility that the partitioning of probe was no longer possible. f cholesterol decreased the formation of LC domains of DPPC, then more of LE phase should have formed. This would imply that cholesterol acted solely on LC phase without affecting LE phase in the co-existence region". Aromicforce microscope (AFM) images Monolayers of DPPC/cholesterol containing 1 wt% of the fluorescent probe NBD-PC were compressed at a rate of 0.04 nrrr' molecule " min- ' to a surface pressure of 7 mn m-, visualized using epifluorescence microscopy, as shown in Fig. SA. The same monolayer was deposited onto mica at a rate of 10 mmmin' by Langmuir-Blodgett transfer technique and analyzed using contact mode of AFM; the height image of a 60 urn x 60 urn area of the LB film is shown in Fig. SB. To the contrary of fluorescence images, brighter regions in the AFM height image corresponded to higher features, while darker regions were indicative of lower features. Higher features resulted from straightening of the hydrophobic tails caused by compression of the monolayer. Thus, in contrast to fluorescence, LC regions appeared brighter than LE regions Both images in Fig. S appeared to be in agreement in terms of size and shape of the domains'r''". Based on this correlation, it is evident that air-water interface of a monolayer can be reproduced onto mica and can, therefore, be visualized by AFM. n AFM height images of LB films with 1% cholesterol at a surface pressure of 7 rnnm- 1 (Fig. 6A), the LC domains were not in any direct contact with one another. However, at 10 rnnm 1 surface pressure, a complex network of LC region was formed that was contiguous throughout the sample (Fig. 6B). This suggested that percolation threshold was between 7 and 10 mnm- l, i.e., cholesterol started fusing the organized DPPC domain structure at this region of surface pressure. Section analysis of each AFM height image was performed (Fig. not shown). At surface pressures of 7 and 10 mnm-', the average heights of the LC domains were measured to be 1.1 ± 0.2 and 1.1±0.1 nm, respectively. Fig. 7 shows AFM height images for the DPPC/cholesterol mixture containing 3% cholesterol and 1 wt% of NBD-PC. mage A contains LC regions in a variety of shapes. From the section analysis of this image (not shown), average height difference between LC and LE regions was measured to be Fig. 5--A monolayer of DPPC:cholesterol (98: ) containing 1 wt% of the fluorescent probe NBD-PC compressed at a rate of 0.04 nm 2 molecule- l min- 1 to a surface pressure of7 mnm-'. A: 60 urn x 60 urn image of the monolayer visualized using epifluorescence microscopy. The same monolayer deposited onto mica at a rate of 10 mrnmin and analyzed using contact mode of AFM. B: An AFM height image of a 60 urn x 60 um area of the LB film. The average size of the LC domains in the fluorescent image, A, 29 )lm 2 and AFM height image, B, was calculated to be 30 )lm 2 o 10 um um 10 Fig. 6-AFM height images of a monolayer of DPPC:cholesterol (98:1) containing 1 wt % NBD-PC at surface pressures of 7 and 10 mnm-' [A: 7 mn rn'"; B: 10 mnm-l]

6 PANDA et al.: MORPHOLOGY OF MXED DPPC-CHOLESTEROL FLMS 119.l±O.2 nm. mage B contains a large area of spiral LC regions stemming from shapes of LC regions as seenin image A. The section analysis of this image measuredan average height difference of 0.9±0.2 nm betweenlc and LE regions. An area of 1 urn x 1 urn from LB films was analyzedusing AFM for mixed monolayer containing 6% cholesterol (Fig. 8). mage A shown in Fig. 8 containstwo phases, as shown by presence of both bright(lc) and dark (LE) areas. The section analysis of image A and B measured an average height differenceof 1.1±0.1 and 1.1±0.05 nm between the tworegions. Although practically no domain was visible in the epifluorescence microscopy, for samples with 10% cholesterol,afm height images (Fig. 9) evidenced twophases, being present in the monolayer. t is clear from the figure that a fractal design had been originatedboth in A and B. During the formation of LC regions, cholesterol, as it is known to have a fluidizingproperty, decreased the line tension of the LC-LE boundary, for which such fractal designs appeared. The section analyses of these images measured an average height difference of 1.04 ± 0.01 and 1.09 ± 0.01 nm between the LC and LE regions, respecti vel y. When a detailed cross-section analysis was performed on a LB deposit of mixed monolayer of DPPC/cholesterol containing 3% cholesterol, then the existence of spiral shaped regions were evidenced. Fig. 10 is a cross-section analysis of a zoomed region of Fig. 7B. Here, typical height difference was in consistent with height for cholesterol. Moreover, the o 51lln J.ln nm 5.0 OM 0.0 OM Fig. 9--AFM height images of a monolayer of DPPC:cholesterol (89: 10) containing 1 wt% NBD-PC at surface pressures of 7 and 10 mnm-' [A: 7 mnm-'; B: 10 mnm-'] OM 6.0 OM o J.1n OM Fig.7-AFM height images of a monolayer of DPPC:cholesterol (96:3) containing 1 wt% NBD-PC at surface pressures of 7 and 10 mnmoira:7 mnm-'; B: 10 mnm"] 3.0 OM Oflill ~TB o 1.5 OM 0.51lln OM Fig.8-AFM height images of a monolayer of DPPC:cholesterol (93:6) containing 1 wt% NBD-PC at surface pressures of 7 and 10 mnm-'[a:7 mnm-'; B: 10 mnm-'] 111. Fig. lo--afm image cross-section analysis of DPPC:cholesterol (96:3) containing wt% NBD-PC at surface pressure of 10 mnm-' [A: AFM height image; B: cross-section analysis of the portion mentioned by arrows in A] ZOo

7 120 NDAN J. BOCHEM. BOPHYS., VOL. 40, APRL 2003 ups and downs follow a systematic order, which is of typical height for the cholesteroli". t clearly indicates the edges of the spirals to be enriched with cholesterol. The enhanced resolution provided by AFM images showed that two phases were consistently present (Figs 6-10). Both bright and dark areas in the AFM images were apparent as the amount of cholesterol was increased from 1 to 10%. There was a consistent height difference of approx. 1 nm between LC and LE phases, corresponding to the accepted height difference between such phases':". From the analyses of both the pressure-area isotherms and AFM images, it is clear that cholesterol did not solely convert LC to LE phase. Therefore, in monolayers containing 6% and 10% cholesterol, the fluorescent probe must not be properly reporting the presence of LC phase in the fluorescence images. Without the enhanced resolution, which the AFM provides for visualizing mono layers in the co-existence region, it would not have been possible to confirm the presence of both phases. The fluorescence mage revealed that 1% cholesterol caused a decrease in line tension at a surface pressure of 10 mlxm' (Fig. 2 B, C and D). These spiral structures seen in fluorescence image of the DPPC monolayer containing 1% cholesterol were not visible when the same monolayer was deposited and visualized AFM (Fig. 6). On the contrary, the presence of these spiral structures in AFM images illustrated that 3% cholesterol caused a decrease in line tension at a surface pressure of 10 mnm- 1 (Fig. 7B). Fluorescence microscopy of the monolayer, before it was deposited did not reveal these spiral structures (Fig. 4 A, D). These observations suggest that when a monolayer is deposited onto mica, the monolayer surface pressure may not be the same as the pressure at the air-water interface. Sometimes, if the material transferred to mica surface is not compensated by film compression, there is a possibility of anomalies. But, in our case we took utmost care about it. During the progress of time, there is also a possibility of spreading out of the film along the mica surface. Secondly, drying up of the deposit with time also plays a crucial role. n order to reproduce spiral LC domains at the air-water interface, it was necessary to increase the amount of cholesterol from 1% to 3%. t, therefore, may be required to exercise caution when attempting to deposit an air-water interface onto a solid substrate. The visualization of spiral LC domains by AFM demonstrated that these structures were real and not a probe induced artifact seen in fluorescence microscopy. Also, it proved that spirals were stable features since no drastic change occurred over time. The sharp invaginations between spirals may possibly indicate the presence of enriched cholesterol in such regions since the spirals did not dissolve when they were in very close proximity (Fig. 10). This was in agreement with the observation that cholesterol gets enriched in the phase boundary". The fluctuation of height changes between LC and LE regions of the spirals shown in Fig. 10, simply could not be explained. An initial explanation was dehydration of the sample, which could be used to explain the height difference of the two regions of the spirals from +0.6 nm to nm. Conclusion Surface pressure-area measurements, epifluorescence microscopy and AFM were not independently effective in studying the co-existence of both LE and LC region in a monolayer, rather a combination of these was found to be more useful. Surface pressurearea measurements helped in identifying the condition for phase transition, midpoint of phase transition and compressibilities. Epifluorescence microscopy was found to be useful in identifying the LE or LC regions. Due to probe partitioning inconsistency, sometimes LE and LC regions were found to be indistinguishable. AFM could provide a higher resolution of the co-existence region when discrepancies occurred in fluorescence measurements. Phase transition of pure DPPC gradually disappeared with progressive addition of cholesterol, while the pressure at the midpoint remained unchanged. Even the compressibility of the DPPC film did not change significantly with increasing amount of cholesterol, possibly due to the property of cholesterol in fluidizing LC and condensing LE region of the films. ncreasing amount of cholesterol in mixed monolayers resulted in gradual decrease in LC region, which appeared to be homogeneous at 10 wt% cholesterol. Fluidizing property of cholesterol and partitioning of the probe into both LE and LC regions contributed such phenomena. Presence of both phases was confirmed by AFM measurements. The above results indicate that although cholesterol fluidized LC region of DPPC films, it could not solely convert it into LE phase.

8 PANDA et al.: MORPHOLOGY OF MXED DPPC-CHOLESTEROL FLMS 121 Acknowledgement The work has been supported by grants received fromthe National Science and Engineering Research Council(NSERC), Canada, A K Panda gratefully acknowledges the receipt of the BOYSCAST Fellowship from the Department of Science & Technology,Govt. of ndia, New Delhi, for pursuing theresearch at the University of Western Ontario. References Creuwels L A J M, vangolde L M G & Haagsman H P Lung (1997) 175, 1-39 Pattie R E J Reprod Ferti! Suppl (1975) 23, Nag K, Harbottle R R & Panda A K J Surf Sci Technol (2000) 16, Veldhuizen R A, Nag K, Orgeig S & Possmayer F Biochim Biophys Acta (1998) 1408, Yu S H & Possmayer F, J Lipid Res (2001) 42, Orgeig S & Daniels C B Comp Biochem Physiol A Mol lntegr Physiol (2001) 129, Yeagle P L Biochim Biophys Acta (1985) 822, Yeagle P L Cholesterol Membr Models (1993) 1-12 Worthman L-A D, Nag K, Davis P J & Keough K M Biophy 1(1997)72, Ahmed S N, Brown D A & London E Biochemistry (1997) 36, Mueller-Landau F & Cadenhead D A Chem Phys Lipids (1979) 25, Mueller-Landau F & Cadenhead D A Chem Phys Lipids (1979) 25, Evans R W, Williams M A & Tinoco J Biochem J (1987) 245, Chou T-H & Chang C-H Colloids Surf B Biointerfaces (2000) 17, Serfiss A B, Brancato S & Fliesler S J Biochem Biophys Acta (2001) 1511, Panda A K, Nag K, Harbottle R R, Possmayer F & Petersen N 0 J Colloid nterface Sci (Submitted) 17 Radhakrishnan A, Li X-M, Brown R E & McConnel H M Biochim Biophys Acta (2001) 1511, Nag K, Munro J G, Hearn S A, Rasmusson J, Petersen N 0 & Possmayer F J Struct Bioi (1999) 126, Ding J, Takamoto D Y, von Nahmen A, Lipp M M, Lee K Y C, Waring A J & Zasadzinski J A Biophys J (2001) 80, Davis J T & Rideal E K in nterfacial phenomena (1963), Academic Press, New York, pp Gaub H E, Moy V T & McConnel H M J Phys Chem (1986) 90, Panda A K, McConologue C W, Munro J G, Harbottle R R, Konermann L, Petersen N 0, Vanderlick T K, Possmayer F & Nag K J Str Bioi (Submitted)