Review

Transmembrane Asymmetry and Lateral Domains inBiological Membranes

Philippe F. Devaux1,* and Roger Morris2

1 Institut de Biologie Physico-Chimique, 13 rue Pierre etMarie 75005 Paris, France, 2MRC Centre forDevelopmental Neurobiology, King’s College, LondonSE1 1UL, UK*Corresponding author: Philippe F. Devaux,Philippe.Devaux@ibpc.fr

It is generally assumed that rafts exist in both theexternal and internal leaflets of the membrane, and thatthey overlap so that they are coupled functionally andstructurally. However, the two monolayers of the plasmamembrane of eukaryotic cells have different chemicalcompositions. This out-of-equilibrium situation is main-tained by the activity of lipid translocases, which com-pensate for the slow spontaneous transverse diffusion oflipids. Thus rafts in the outer leaflet, corresponding todomains enriched in sphingomyelin and cholesterol, can-not be mirrored in the inner cytoplasmic leaflet. Theextent to which lipids contribute to raft properties canbe conveniently studied in giant unilamellar vesicles. Inthese, cholesterol can be seen to condense with satur-ated sphingolipids or phosphatidylcholine to form mmscale domains. However, such rafts fail to model bio-logical rafts because they are symmetric, and becausetheir membranes lack the mechanism that establishesthis asymmetry, namely proteins. Biological rafts are ingeneral of nm scale, and almost certainly differ in size andstability in inner and outer monolayers. Any couplingbetween rafts in the two leaflets, should it occur, is prob-ably transient and dependent not upon the properties oflipids, but on transmembrane proteins within the rafts.

Key words: lipid asymmetry, lipid protein interactions,liquid ordered phase, rafts, transmembrane diffusion

Received 12 January 2004, revised and accepted for pub-lication 19 January 2004

The concept of lateral segregation in biological membranes

developed over the last 30 years as a purely thermody-

namic description by physicists to explain the coexistence

within a bilayer of more than one lipid phase. The lipid

phases thought to correspond to biological membranes

are sketched in Figure 1: La (also called Ld for liquid dis-

order), the ‘normal’ fluid phase; Lb, the nonphysiological

ordered gel phase formed by cooling membranes; and the

LO (liquid-ordered) phase in which the lipids are ordered (as

in Lb) but retain their freer rotational and lateral diffusion

(as in La) and which can occur at physiological tempera-

tures (1). Bilayers in model systems are generally symmet-

rical with identical lipids in the outer and inner monolayers

(1). However, these phases can separate out in the same

bilayer to give different lateral domains. The coexistence of

several phases in Giant Unilamellar Vesicles (GUVs) made

from mixtures of lipids (generally 1–4 components) can be

observed and accounted for within the framework of clas-

sical thermodynamics (2).

The potential biological advantages of forming different

lateral domains within a membrane were emphasized

many years ago by McConnell (3). However, biological

membranes are not at equilibrium, and in particular maintain

an asymmetry in lipid and protein composition between

their inner and outer leaflets. These critical factors are not

part of the physicists’ description of lamellar phase separ-

ation, and significantly alter how we should think of the

structure and organisation of biological rafts. Here we con-

sider the physical consequences of the difference between

biological membranes and the simple models used by

physicists, particularly those arising from lipid asymmetry.

Can Lipids in the Different Leaflets of a Bilayerbe Maintained at Equilibrium in DifferentPhases?

Since asymmetry between the outer and inner mono-

layers, and divergence from thermal equilibrium, are two

major differences between biological and model mem-

branes, it is worth first asking whether asymmetric lamel-

lar phases can exist at thermodynamic equilibrium.

The answer is yes. Luzzati and collaborators (4–6)

described several phases consisting of mixed bilayers

with one monolayer of type a (disordered chains) and the

other of type b (ordered chains). Figure 2a shows a phase

reported for egg lecithin or phosphatidic acid in water

denominated Lab, which consists of disordered mosaics

of two types of domains where monolayers a and b can be

randomly associated (4). Bilayers of Lg phase of type II

obtained with mitochondrial lipids are also mixed with one

monolayer of type a and one monolayer of type b (5). Thus

asymmetric lateral domains, with alternative fluid and

ordered regions, can be formed even with a single lipid

component at thermal equilibrium. Proteins can also trig-

ger an asymmetrical bilayer in model systems. For example,

myelin basic protein on one side of a lipid bilayer

Traffic 2004; 5: 241–246Copyright # Blackwell Munksgaard 2004

Blackwell Munksgaard doi: 10.1111/j.1600-0854.2004.00170.x

241

(composed of phospholipids and sulfatides) can modify

selectively the phase of the monolayer interacting with it

(6), hence creating an asymmetric bilayer (Figure 2b).

Sonication of a mixture of lipids with head groups of different

size (phosphatidylethanolamine, PE, and phosphatidylcho-

line, PC) also leads to stable asymmetric bilayers (7). The

high curvature of small sonicated vesicles is the driving force

for such segregation but this is not a real thermal equilibrium.

Thus it is possible in principle to have two adjacent mono-

layers that differ in their lipid organization; it is not necessary

that the lipids in both leaflets of rafts be in the same phase.

What is Known About Transmembrane LipidAsymmetry in the Membrane of EukaryoticCells?

In the plasma membrane of eukaryotic cells aminophos-

pholipids (phosphatidylserine and phosphatidyletholamine)

are predominantly exposed on the cytosolic leaflet,

whereas phosphatidylcholine and sphingomyelin are

predominantly if not exclusively located on the outer

leaflet. Minor lipids such as phosphoinositol and phos-

phatidic acid are predominantly in the inner leaflet,

whereas glycosphingolipids face the outer surface. This

scheme is apparently a ubiquitous property of animal cells,

and probably of plant cells as well (although the relative

proportions of glycerolipids and sphingolipids differ).

This asymmetric distribution is established in post-Endo-

plasmic Reticulum (ER) compartments, for sphingolipids

during biosynthesis in the Golgi (8) and for glycerolipids by

the action of ATPases such as the aminophospholipid trans-

locase that transports aminophospholipids from the outer to

the inner leaflet of plasma membranes (9), and MDR pro-

teins that transport PC in the opposite direction. Although

ATP-independent flippases rapidly randomize lipid composi-

tion across the membrane in the ER, and possibly some

other intracellular membranes, such proteins appear to be

absent from the plasma membrane in which there is only a

very slow (hours) intrinsic diffusion (flip/flop) of glycero- and

sphingolipids between the leaflets of the bilayer that is

opposed by the action of lipid-transporting ATPases (8,9).

The transmembrane distribution of fatty acids is hard to

determine experimentally, and may vary considerably from

one cell type to another. In the erythrocyte plasma mem-

brane, which is the most studied membrane, the inner

leaflet contains more unsaturated lipids than the outer

leaflet. Cholesterol distribution between the bilayers is

unclear. Several reports based on data obtained with fluor-

escent (dehydroergosterol) or spin-labeled cholesterol ana-

logs suggest that cholesterol diffuses spontaneously from

one leaflet to the other with a half-time of the order of a

few minutes at 4 �C (10,11). A flip/flop faster than 1 s was

reported for natural cholesterol (12). However, the steady

state transmembrane distribution will be determined not

just by the exchange rate between bilayers, but also by the

affinity of cholesterol for the other lipid and protein com-

ponents of the membrane which are not distributed sym-

metrically. Asymmetric transmembrane distributions of

cholesterol have been reported for a variety of cell types

and with a variety of techniques (1,13), including preferen-

tial presence (75%) in the inner rather than the outer leaflet

of erythrocytes (10). It is clear that cholesterol can be in

either monolayer, but should not be assumed to be

distributed evenly across the bilayer.

The local viscosity of inner and outer leaflets of eukaryotic

plasma membranes differ. Electron spin resonance (ESR)

with spin-labeled lipids (14) as well as lateral diffusion

measurements carried out by photobleaching with fluores-

cent lipids indicated that the cytosolic leaflet of human

erythrocytes was more fluid than the exoplasmic leaflet

(15). Analysis in model systems with liposomes made with

lipids of the inner or outer leaflet revealed that the differ-

ence in viscosity could not be accounted for by a differ-

ence in cholesterol content, but more likely was due to the

difference in fatty acid chain composition of the two leaf-

lets (16). Experiments with human fibroblasts (17) and

with endothelial cells (18) also indicated a higher mobility

in the inner leaflet lipids. In such experiments one cannot

be sure that the probes report the viscosity of both raft and

nonraft membranes. However, it remains a fact that the

two monolayers have different physical properties, and the

inner leaflet has an average lower viscosity than the outer.

In summary, both the composition and physical properties

of lipids in each monolayer of biological membranes are

Figure 1: Classical symmetrical

lamellar lipid phases. From left

to right: the fluid La phase, the

liquid ordered LO, and ordered Lb

gel phase. Adapted from Ref. 1.

Devaux and Morris

242 Traffic 2004; 5: 241–246

different. Although asymmetry of physical properties in the

two monolayers can be maintained at thermal equilibrium,

asymmetry in chemical composition cannot. Proteins must

be involved in maintaining the asymmetry in composition,

which raises the difficult question: do proteins chose their

preferred lipid environment among the existing phases, or

do proteins organize their lipid microenvironment? If the

latter, the lipid ‘phase’ is not a pure lipid system but is

defined by, and exists as, a lipid-protein domain.

How relevant is lipid phase separation tomembrane rafts?

With living organisms it is difficult to decide what is the

consequence of the physical chemistry of a selected pool

of molecules and what is due to ongoing protein activity,

hence to a sequence of events programmed in the gen-

ome and maintained out of equilibrium by cellular activity.

Typically, mixtures of lipids that are allowed to swell in an

aqueous environment form liposomes at thermodynamic

equilibrium. This does not automatically imply a single

phase. A variety of experiments have shown the coexis-

tence of lipid phases in model systems composed of lipid

mixtures with various head groups and alkyl chains. Unlike

the simple systems studied by physicists, eukaryotic cell

membranes are composed of several hundred chemically

different lipids, differences that can profoundly affect their

behavior in a bilayer (1). It is not conceivable to draw a

rigorous and comprehensive phase diagram of so many

lipids. Nevertheless, the lipids of biological membranes

show striking resemblances to phases described in

model systems in terms of lipid density, bilayer spacing,

chain dynamics, rates of lateral and transverse diffusion.

In spite of those similarities, there are at least three ser-

ious differences between lateral domains in biological

membranes and in giant liposomes. The first is transmem-

brane symmetry� in model systems such as GUVs the

two leaflets of the bilayer are identical in composition.

Phase separation to form discrete LO domains occurs

symmetrically in both leaflets, suggesting a strong coup-

ling between monolayers in these model membranes that

may not occur in asymmetric biological membranes.

The second difference concerns the size and lifetime of

lateral domains. Lateral domains in liposomes are mm scale

and can be readily seen by light microscopy (e.g. two-

photon fluorescence microscopy) where they are generally

observed to be stable for at least minutes (19–22). There is

no current agreement on the size and stability of biological

rafts, reflecting the diversity of rafts that exist even within

a single cell (1); however, it is clear they lie below the

resolution of conventional microscopy (sub 200 nm).

Methods that detect rafts on cells within the 200 nm

level of resolution (deconvolved fluorescence microscopy,

electron microscopy (EM), single/multiple particle tracking)

in general report a diversity of size and shape within the

10–300 nm size range, and stability at least in the range of

seconds (1,23–25). However, others using single particle

tracking and ESR (e.g. 26) describe much smaller (a few

molecules) and more labile (half-lives of ns–ms) rafts, that

may coalesce into larger and more stable structures in

Figures 2: Asymmetrical lamellar structures of lipid

suspensions in water indicating that an ordered phase on

one monolayer does not automatically imply an ordered

phase on the monolayer to which it is coupled. a) Schematic

view of the structure of the lipid lamellae of the Lab phase. The

black dots represent the polar groups, the wiggly lines the

hydrocarbon chains in the a conformation, the straight lines the

chains in the b conformation. The dotted straight lines are the

boundaries of the lipid leaflets. A and B represent lamellae of Lab.

The left hand portion shows the case of monolayers of type a or bapposed two by two via their hydrophobic faces. The right hand

portion shows a distribution of a and b randomly associated

(adapted from (4)). b) The structure on the right shows

asymmetrical lipid bilayers formed in the presence of myelin

basic protein (reproduced from (6)).

Rafts and Transmembrane Asymmetry

Traffic 2004; 5: 241–246 243

response to signaling (26,27). It is worth asking whether

the half-life of residence of individual components (particu-

larly probes with non-native structures) within a raft accu-

rately reflects the stability of the raft as a whole. To borrow

an analogy of a very different two-phase system, a cloud is

described microscopically as a collection of labile microdo-

mains of liquid water concentrated within a gaseous phase

matrix; on a different scale, a larger and more stable object

is seen that differs markedly in properties from surround-

ing dry air. Rafts also may be concentrations of nm lateral

domains within particular regions of the membrane that

impart different properties to 10–200 nm scale regions

(more commonly called rafts) as a whole.

The third, and possibly decisive, difference between

protein-free models and biological membranes is that the

latter contain proteins, which would influence raft compos-

ition, sizes, shapes and overall physical properties, to some

extent independently of thermodynamic considerations of

the pure lipid phases (28–30). For example, selective lipid

protein interactions can be responsible for a particular cloud

of lipids surrounding a given protein, even if rapid exchange

of lipids takes place at the boundary of this protein (28,29).

Since there are no membrane proteins that are symmet-

rical, its shell of ‘boundary’ lipids will not be the same in

each leaflet. Note that proteins can also be responsible for

lateral domains in biomembranes, via a different mechanism.

Indeed, rafts are often viewed as restricted areas of free

lateral diffusion limited by a protein barrier, for example a

cytoskeletal meshwork (25,31). The role of the cytoskeleton

raises the question of the sidedness of this interaction

since the cytoskeleton is situated in the cell cytoplasm.

A limitation of the free diffusion in the outer monolayer is

understandable only for transmembrane proteins, whereas

many of the better characterized external raft proteins are

lipid (glycosylphosphatidyl inositol, GPI)-anchored and could

only link to the cytoskeleton via an intermediate trans-

membrane protein. In summary, biological rafts are smaller,

more diverse and much more complex than the lateral

domains formed in model membranes. Raft size, shape

and composition are likely to be determined by metabolic

rather than thermodynamic considerations.

The Role of Transmembrane Proteins andLipid–Protein Affinities in Coupling BothLeaflets: Hypotheses and Facts

Biological rafts cannot be preformed symmetrical domains

in both leaflets of the bilayer, each containing a high ratio

of sphingomyelin and cholesterol in the L0 phase, into

which selective proteins of the extracytoplasmic or cyto-

solic leaflets converge and couple to each other to form

strategic platforms for signaling, trafficking and the like.

However, it is clear from the biology that rafts exist in both

leaflets of the bilayer (1). There is also considerable evi-

dence from model systems that cholesterol can form con-

densed, ordered phases in the PE/PS (phosphatidylserine)

dominated inner leaflet just as it does in the PC/sphingolipid

dominated outer leaflet, provided these lipids are saturated

or any unsaturation is restricted (as it normally is in higher

organisms) to the sn-2 position (see (1), and references

cited therein (32,33)). There is no theoretical problem with

the existence of rafts in both leaflets of the bilayer,

although the physical properties of these rafts must differ,

reflecting their different composition and possibly size.

Several publications have reported that inner layer rafts

comprising PC and cholesterol or PE and cholesterol

would be less stable than outer leaflet rafts containing

sphingomyelin and cholesterol (26,34). Furthermore it has

been reported that L0 domains formed with PE and cho-

lesterol or PS and cholesterol in models systems have

different shapes than domains obtained with sphingomye-

lin and cholesterol (35). Thus rafts in the outer and inner

leaflets leaflet of biomembranes do not match each other.

However, are rafts in the two leaflets coupled? Of the

properties of lipids that could couple across the bilayer,

interdigitation of long (C24) fatty acid chains, and/or oscilla-

tion of cholesterol in the vertical plane, where it penetrates

up to one third of the adjoining leaflet, have been sug-

gested (1). The contribution of neither of these mechan-

isms to the coupling of leaflets has been investigated

experimentally. A simpler possibility lies in the relationship

of membrane curvature to density: a 1% mismatch in lipid

density gives rise to large effects on liposome shapes (36).

LO phases are condensed, that is, they have a higher

density of lipid than the corresponding La phase. Unless

they are matched in both leaflets of the bilayer they will

curve the membrane (37). Where such curvature is not

allowed (e.g. because of stronger forces from the cytoskel-

eton), rafts in each layer would be forced to be coupled by

the shape of the membrane.

However, at least in some cases, the two leaflets have to

be coupled by transmembrane proteins. A puzzling obser-

vation is that the classical examples of transmembrane

interaction between two proteins found in rafts, which

are cited as proofs of their transmembrane character,

involve a GPI-anchored protein in the outer leaflet and a

diacylated kinase in the inner leaflet (38,39). Both proteins

are peripheral proteins that bind to membranes only

because of their covalent link to membrane lipids

embedded in the lipid phase. How the two proteins inter-

act through the membrane, and how physiological informa-

tion is transferred from one side to the other, is a mystery,

unless a transmembrane partner is also involved (40,41).

In their recent review, Subczynski & Kusumi (26) summar-

ized their earlier work by proposing a rather elaborate

model of rafts in outer and inner leaflets. This model

does not suggest that the properties of the outer leaflet

are mirrored in the inner, as sometimes implied. Without

going into a detailed analysis of their model, it suffices to

say that their major conclusions are against the idea of

Devaux and Morris

244 Traffic 2004; 5: 241–246

symmetrical rafts. The sizes and lifetimes of ‘rafts’ on the

two sides of biomembranes would be very different and

would allow only brief cross interaction. The outer leaflet

would have small unstable rafts, containing only a few

proteins, with a lifetime in the millisecond range. But

these small rafts could form more stable larger domains

after receptor activation and oligomerization. The lifetime

of the latter rafts would be of the order of a few minutes.

The inner leaflet, on the other hand, would have only small

unstable rafts with a very short lifetime, each containing

perhaps one molecule of the downstream effector mol-

ecules. Figure 3 schematizes the differences between the

two types of rafts possibly encountered on the two sides

of the biomembranes. It emphasizes the difference in time

scale and size but indicates that cross-linking can take

place via transmembrane proteins.

Conclusions

Because the size of rafts in biological membranes is below

the resolution of the optical microscope and because inner

membrane rafts are ill-defined small entities that coalesce

with each other or reform continuously, it will remain

difficult to demonstrate to what extent rafts in the two

leaflets match each other. Further work is needed on

protein and lipid localization with particular attention to

the size and time scale of the domains. Already, it appears

impossible to envisage rafts as permanent patches of

symmetrical domains of lipid-ordered phase. Even if rafts

in both sides of a membrane are superimposed for short

periods, which may have important physiological func-

tions, this connection has a short lifetime.

Acknowledgments

The authors would like to thank Dr. V. Luzzati for bringing their attention to

his former publications, that demonstrated the existence of asymmetrical

lamellar lipid phases. The authors acknowledge support from the EU HPRN-

CT-2000–00077 (research training network on sphingolipid synthesis and

organization). P.F.D. was supported by grants from the Centre National de

la Recherche Scientifique (UMR CNRS 7099).

Figure 3: Schematic representation of biomembrane rafts emphasizing the asymmetry of the rafts in the twomonolayers, which

have different sizes (and probably different lifetimes). Both monolayers have liquid-ordered phases with cholesterol but the

phospholipids interacting with cholesterol are different and hence induce slightly different LO phases. This figure suggests that

superposition of rafts in two monolayers requires a coupling via a transmembrane protein. The association can be also fortuitous and is

likely to be temporary.

Rafts and Transmembrane Asymmetry

Traffic 2004; 5: 241–246 245

References

1. Morrris R, Cox H, Mombelli E, Quinn PJ. In: Quinn PJ, ed. Membrane

Dynamics and Domains. London: Kluwer Academic Plenum Publishers,

2004 p. 35–118.

2. Almeida RFM, Fedorov A, Prieto M. Sphingomyelin, phosphatidyl

choline, cholesterol phase diagram: boundaries and composition of

lipid rafts. Biophys J 2003;85:2406–2416.

3. McConnell HM. Structures and transitions in lipid monolayers at the

air–water interface. Annu Rev Phys Chem 1991;42:171–195.

4. Ranck JL, Mateu L, Sadler DM, Tardieu A, Gulik-Krzywicki T, Luzzati V.

Order-disorder conformational transitions of the hydrocarbon chains of

lipids. J Mol Biol 1974;85:249–277.

5. Ranck JL, Zacca G, Luzzati V. The structure of a lipid-water lamellar

phase containing two types of lipid monolayers. An X-ray and neutron

scattering study. J Appl Cryst 1980;13:505–512.

6. Mateu L, Luzzati V, London Y, Gould RM, Vosseberg FGA, Olive J.

X-ray diffraction and electron microscopy study of the interactions of

myelin components. The structure of a lamellar phase with a 50–180 A

repeat distance containing basic proteins and acidic lipids. J Mol Biol

1973;75:697–709.

7. Op den Kamp JAF. Lipid asymmetry in biomembranes. Annu Rev

Biochem 1979;48:47–71.

8. Raggers RJ, Pomorski T, Holythius JCMK, Kalin N, van Meer G. Lipid

traffic: the ABC of transbilayer movement. Traffic 2000;1:226–234.

9. Devaux PF. Static and dynamic lipid asymmetry in cell membranes.

Biochemistry 1991;30:1163–1173.

10. Schroeder F, Nemecz G, Wood WG, Joiner C, Morrot G, Ayrault-Jarrier M,

Devaux PF. Transmembrane distribution of sterol in the human

erythrocyte. Biochim Biophys Acta 1991;1066:183–192.

11. Muller P, Herrmann A. Rapid transmembrane movement of spin-

labeled steroids in human erythrocytes. Biophys J 2002;82:1418–1428.

12. Steck TLYeJ, Lange Y. Probing red cell membrane cholesterol move-

ment with cyclodextrin. Biophys J 2002;83:2118–2125.

13. McIntosh TJ. Organization of skin stratum corneum extracellular lamel-

lae: diffraction evidence for asymmetric distribution of cholesterol.

Biophys J 2003;85:1675–1681.

14. Seigneuret M, Zachowski A, Hermann A, Devaux PF. Asymmetric lipid

fluidity in human erythrocyte membrane: new spin-label evidence.

Biochemistry 1984;23:4271–4275.

15. Morrot G, Cribier S, Devaux PF, Geldwerth D, Davoust J, Bureau JF,

Fellmann P, Herve P, Frilley B. Asymmetric lateral mobility of phospho-

lipids in the human erythrocyte membrane. Proc Natl Acad Sci U S A

1986;83:6863–6867.

16. Cribier S, Morrot G, Neumann JM, Devaux PF. Lateral diffusion of

erythrocyte phospholipids in model membranes comparison between

inner and outer leaflet components. Eur Biophys J 1990;18:33–41.

17. El Hage Chahine M, Cribier S, Devaux PF. Phospholipid transmem-

brane domains and lateral diffusion in fibroblasts. Proc Natl Acad Sci

U S A 1993;90:447–451.

18. Julien M, Tournier JF, Tocanne JF. Differences in the transbilayer and

lateral motions of fluorescent analogs of phosphatidylcholine and phos-

phatidylethanolamine in the apical plasma membrane of bovine aortic

endothelial cells. Exp Cell Res 1993;208:387–397.

19. Bagatoli LA, Gratton EA. Two photon fluorescence microscopy of

coexisting lipid domains in giant unilamellar vesicles of binary phospho-

lipids mixtures. Biophys J 2000;78:290–305.

20. Bagatoli LA, Gratton EA. Correlation between lipid domain shape and

binary phospholipid mixture composition in free standing bilayers: a

two photon fluorescence. Biophys J 2000;79:434–447.

21. Dietrich C, Bagatoli LA, Volovyk ZN, Thompson NL, Levi M, Jacobson K,

Gratton E. Lipid rafts reconstituted in model model membranes.

Biophys J 2001;80:1417–1428.

22. Kahya N, Scherfeld D, Bacia K, Poolman B, Schwille P. Probing

lipid mobility of raft-exhibiting model membranes by fluorescence

correlation spectroscopy. J Biol Chem 2003;278:28109–28115.

23. Gaus K, Gratton E, Kable EP, Jones AS, Gelissen I, Kritharides L,

Jessup W. Visualizing lipid structure and raft domains in living cells

with two-photon microscopy. Proc Natl Acad Sci U S A 2003;

100:15554–15559.

24. Schutz GJ, Kada G, Pastushenko VP, Schindler H. Properties of lipid

microdomains in a muscle cell membrane visualized by single molecule

microscopy. EMBO J 2000;19:892–901.

25. Dietrich C, Yang B, Fujiwara T, Kusumi A, Jacobson K. Relationship of

lipid rafts to transient confinement zones detected by single particle

tracking. Biophys J 2002;82:274–284.

26. Subczynski WK, Kusumi A. Dynamics of raft molecules in the cell

and artificial membranes. approaches by pulse EPR spin labeling

and single molecule optical microscopy Biochim Biophys Acta

2003;1610: 231–243.

27. Simons K, Toomre D. Lipid rafts and signal transduction. Nature Rev

Mol Cell Biol 2000;1:31–39.

28. Devaux PF, Seigneuret M. Specificity of lipid–protein interactions as

determined by spectroscopic techniques. Biochim Biophys Acta

1985;822:63–125.

29. Marsh D, Horvath LI. Structure, dynamics and composition of the

lipid–protein interface. Perspectives from spin-labelling. Biochim Bio-

phys Acta 1998;1376:267–296.

30. Bloom M, Thewalt JL. Time and distance scales of membrane domain

organization. Mol Membr Biol 1995;12:9–13.

31. Suzuki K, Sheetz MP. Binding of cross-linked glycosylphos-

phatidylinositol-anchored proteins to discrete actin-associated

sites and cholesterol-dependent domains Biophys J 2001;81:

2181–2189.

32. Wang TY, Silvius JR. Cholesterol does not induce segregation of

liquid-ordered domains in bilayers modeling the inner leaflet of the

plasma membrane Biophys J 2001;81:2762–2773.

33. Shaikh SR, Brzustowicz MR, Gustafson N, Stillwell W, Wassall SR.

Monounsaturated PE does not phase-separate from the lipid raft mol-

ecules sphingomyelin and cholesterol: role for polyunsaturation? Bio-

chemistry 2002;41:10593–10602.

34. Aussenac T, Tavares M, Dufourc EJ. Cholesterol dynamics in mem-

branes of raft composition: a molecular point of view from 2H and 31P

solid state NMR. Biochemistry 2003;18:1383–1390.

35. Samsonov AV., Mihalyov I, Cohen FS. Characterization of cholesterol

sphingomyelin domains and their dynamics in bilayer membranes.

Biophys J 2001;81:1486–1500.

36. Farge E, PF. Devaux. Shape changes of giant liposomes induced by an

asymmetric transmembrane distribution of phospholipids. Biophys J

1992;61:347–357.

37. Baumgart T, Hess ST, Webb WW. Imaging coexisting fluid domains in

biomembrane models coupling curvature and line tension. Nature

2003;425:821–824.

38. Harder T, Scheiffele P, Verkade P, Simons K. Lipid domain structure of

the plasma membrane revealed by patching of membrane components

J Cell Biol 1998;141:929–942.

39. Holowka D, Sheets ED, Baird B. Interactions between Fce RI and lipid

raft components are regulated by the actin cytoskeleton. J Cell Sci

2000;113:1009–1019.

40. Pyenta PS, Halowka D, Baird B. Cross correlation analysis of inner

leaflet anchored green fluorescent protein co-redistributed with IgE

receptors and outer leaflet lipid raft components. Biophys J

2001;80:2120–2132.

41. Kunzelman-Marche C, Freyssinet JM, Martinez MC. Loss of plasma

membrane phospholipid asymmetry requires raft integrity. J Biol Chem

2002;277:19876–19881.

Devaux and Morris

246 Traffic 2004; 5: 241–246