Modulation of membrane function by cholesterol

Biochimie (1991) 73, 1303-1310 © Soci6t6 frangaise de biochimie et biologie mol6culaire / Elsevier, Paris

1303

Modulation of membrane function by cholesterol PL

Yeagle

Department of Biochemistry, 140 Farber Hall, University at Buffalo (SUNY) School of Medicine, Buffalo, NY 14214 USA (Received 24 May 1991; accepted 17 September 1991)

Summary - - The molecular basis for the essential role of cholesterol in mammalian (and other cholesterol-requiring) cells has long been the object of intense interest. Cholesterol has been found to modulate the function of membrane proteins critical to cellular function. Current literature supports two mechanisms for this modulation. In one mechanism, the requirement of "free volume' by integral membrane proteins for conformational changes as part of their functional cycle is antagonized by the presence of high levels of cholesterol in the membrane. In the other mechanism, the sterol modulates membrane protein function through direct sterol-protein interactions. This mechanism provides an explanation for the stimulation of the activity of important membrane proteins and for the essential requirement of a structurally-specific sterol for cell viability. In some cases, these latter membrane proteins exhibit little or no activity in the absence of the specific sterol required for growth of that cell type. The specific sterol required varies from one cell type to another and is unrelated to the ability of that sterol to affect the bulk properties of the membrane. cholesterol / cholesterol-protein interactions / membrane / cholesterol regulation of enzymes Cholesterol has long been known to be an essential component of m a m m a l i a n cells. Yet despite much study, the role cholesterol plays in m a m m a l i a n cell biology has remained a mystery. Without cholesterol, m a m m a l i a n cells cannot experience normal growth. As a consequence, most m a m m a l i a n cells are capable of making their own cholesterol. M a n y steps are involved in cholesterol biosynthesis. Even after the synthesis of lanosterol, 18 enzymatically catalyzed steps remain to produce cholesterol [1]. Much valuable cellular energy is therefore utilized in the complex biosynthetic pathway to produce the particular chemical structure of cholesterol. W h y this occurs is not fully understood. Bloch has suggested that evolutionary pressure for a more biologically competent sterol led to the development of the pathway from lanosterol to cholesterol [2]. However, the molecular details describing why the cholesterol structure is required for biological competence (for example, in m a m m a l i a n cells) have not yet been fully described.

Structural r e q u i r e m e n t s for sterols in cell biology Some studies are available on specificity for sterol structure by sterol-requiring cells. In vitro experi-

ments have shown that lanosterol cannot fully substitute for cholesterol as the essential sterol for mycoplasma cell function [3]. In particular, Mycoplasma mycoides can be adapted to grow on low cholesterol media. However, they cannot grow in the total absence of cholesterol in the medium, since they do not make their own cholesterol and cholesterol is required for cell growth and function. Supplementation in the medium of lanosterol will not support cell growth in the absence of cholesterol. However, cell growth will occur at nearly the same rate in cells fed low cholesterol levels supplemented with high (relatively) lanosterol levels, as in cells fed high (relatively) cholesterol levels [4]. Thus, cholesterol appears both adequate and necessary for minimal cellular function in these mycoplasma, while for optim u m cellular function higher membrane stero! content is required but w~thout the structural specificity associated with the requirement for minimal cellular function. This sterol synergism has pointed to a special role for cholesterol in supporting cell growth in which the particular chemical structure of cholesterol is required for mycoplasma. The requirement of cholesterol for normal function of m y c o p l a s m a is mirrored in yeast by an analogous requirement for ergosterol [3]. Yet one steroi cannot substitute for the other; cholesterol cannot fully

! 304

PL Yeagle

substitute for ergosterol in yeast, and ergosterol cannot support normal mammalian cellular function. Anaerobic growth of S a c c h a r o m y c e s cerevisiae requires ergosterol supplement to the culture medium. Supplementation only with cholesterol will not support normal growth [5]. Yet cholesterol is more effective at modifying the properties of lipid bilayers than is ergosterol (see below). These data point to a specific sterol recognition interaction, crucial to normal cellular function, that is different in yeast than it is in mycoplasma. The requirement for cholesterol in mammalian cells is not well understood due to the paucity of available mammalian sterol auxotrophes. Therefore, other methods have been used to explore the role of cholesterol. In mammalian cells, inhibition of cellular cholesterol biosynthesis inhibits cell growth [6]. In addition to the role of metabolic products from mevalonate other than cholesterol in cell growth, cholesterol itself played a role since the addition of cholesterol in the culture medium would restore cell growth [7]. Data such as these suggest that a unique sterol structure leads to biological competence in each sterol-requiring organism. What defines that biological competence is central to the discussion in this article. From this analysis emerges an hypothesis for the essential role of cholesterol in mammalian (and other cholesterol-requiring) cells.

Physical effects of cholesterol on membrane structure Cholesterol has been the subject of extensive investigation for many years [8]. Cholesterol is predominantly found in the membranes of cells, due to its largely hydrophobic structure [9]. Much effort has been expended exploring the fascinating effects cholesterol has on the bulk properties of lipid bilayers, including cholesterol effects on permeability [ 10] and molecular ordering [11, 12], lateral phase separations [13], reduction of the enthalpy of phospholipid phase transitions from the gel to the liquid crystalline state [14], to name just a few. As will be seen in the following, this line of study has provided information about some of the effects of cholesterol on biological membrane function (for a review, see [15]). However, the structural (with respect to sterol) specificity of these physical effects does not mirror the considerable structural specificity of some of the biological effects for sterols. And as noted above, some patterns observed in the physical data have no correlation at all with the biological requirement for sterol structures ( i e , t h e yeast data referred to above). Therefore, the search for the role of cholesterol in sterol-requiring cells has broadened to encompass more specific struc-

tural effects that may arise from the interaction between sterols and membrane proteins.

Modulation of biological function of membranes by cholesterol Since the primary location of cholesterol in cells is in the membranes of the cells, the ability of cholesterol to modify membrane function has been the focus of much interest. In particular, the modulation by cholesterol of the function of membrane proteins has been examined, both by modifying the cholesterol content of the nativc membranes in which the protein is found, and by reconstituting the membrane proteins into membranes of defined lipid content. Studies of this sort have led to three different classes of observations: 1) An increase in the level of cholesterol in the membrane leads to a proportionate decrease in membrane protein function. An example can be seen in figure 1. In this example, the equilibrium constant for the Meta I-Meta II transition of rhodopsin was measured as a function of the cholesterol content of reconstituted membranes [16]. The data show an inverse relationship between the function of the membrane protein and the cholesterol content of the membrane. For rhodopsin, cholesterol apparently acts as a negative modulator, or an inhibitor, at high cholesterol levels. Similar inhibitory effects of cholesterol have been observed on rhodopsin activation of the cyclic GMP cascade [17], alkaline phosphatase [18], UDP-glucuronosyltransferase [19], thymidine transport [20], and anion transport [21 ]. 2) An increase in the level of cholesterol in the membrane leads to a proportionate increase in membrane protein function. This ,,,,,"'m r,u,. . . . .~,.,.,, ... ,,~ ,~,.~. . . membrane cholesterol levels for the Na+-K+-ATPase in figure 2. From a cholesterol/phospholipid mol ratio of 0 to about 0.35, an increase in cholesterol in the membrane leads to an increase in the ATP hydrolyzing activity of this enzyme in the modified native membrane. Stimulation of other membrane functions by cholesterol has been observed, including Na+-Ca 2÷ exchange [22], ATP-ADP exchange [23], carder mediated lactate transport [24] and the acetyl choline receptor [25, 26]. (At high membrane cholesterol levels, inhibition is observed as as described in the examples in 1.) 3) Some membrane functions appear to be insensitive to cholesterol levels in the membrane. Figure 3 shows data for the ATPase activity of the rabbit sarcoplasmic reticulum. As membrane cholesterol content was varied over a wide range, no alteration in ATPase activity was observed. Other membrane functions such as sucrase, lactase and maltase activities of the rat intestinal microvillus are apparently unaffected by alterations in the membrane cholesterol level [18].

Essential role of cholesterol in cells 0.8

0.6

!

1305

120 lOO

8o E =

0.4

60

E ¢~

40

E

0.2

20

0.0

I

0

"

I

'

I

10 20 30 Mole % Cholesterol

•

Fig 1. Effect of membrane cholesterol on the equilibrium constant for the Meta I-Meta II transition of bovine rhodopsin in reconstituted systems at 20°C as a function of cholesterol content of the membranes (mol % with respect to the total lipid content of the membranes). Data replotted from [ 16].

The challenge is to find an explanation for these varied observations. The approach to be taken here is to identify the known effects of cholesterol on membrane structure and to use these properties to explain the observed effects of cholesterol on m e m brane function. A review of the available literature indicates that there are at least two general classes of interactions in which cholesterol can engage while in a membrane.

I

•

10

40

I

•

20

I

30

"

I

"

40

I

50

"

60

M o l e % cholesterol

Fig 2. Effect of membrane cholesterol on the ATPase activity of the bovine kidney Na+-K+-ATPase. Cholesterol content was modified by incubation of the native membranes with lipid vesicles, with or without cholesterol, which resulted in intermembrane cholesterol transfer and alterations in the cholesterol level in the native membrane. Data replotted from [42].

[,., <

10

~'~

oml

Sterol modulation of bulk membrane properties The most studied is the class of bulk effects on the lipid bilayer of the membrane. An example of this can be seen in the reduction of passive permeability of a membrane to neutral solutes upon the addition of cholesterol [ 10]. In phosphatidylcholine bilayers, cholesterol is the most effective sterol at reducing passive permeability. Cholesterol reduces permeability in direct proportion to the level of cholesterol in the membrane. Cholesterol, at a level of 50 mol percent with respect to the m e m b r a n e phospholipid, virtually eliminates passive glucose permeability in phosphatidylcholine bilayers. On a molecular level, passive permeability has been modelled in terms of packing defects in the lipid bilayer (for a more detailed description see [27]). These packing defects result because of non-cooperative isomerizations of carbon-carbon single bonds such that adjacent chains do not exactly mimic each

[] •

<

0

[] []

O I

I

|

I

10

20

30

Mole percent cholesterol in SR membrane

Fig 3. Effect of membrane cholesterol on the ATPase activity of the rabbit muscle sarcoplasmic reticulum Ca 2+ATPase [50]. Membrane cholesterol levels were modified as described in figure 2. other in conformation [28]. This leads to transient packing defects into which small molecules can fit. By moving from defect to defect, small molecules can transit the lipid bilayer. The smaller the molecule the more rapidly the transmembrane movement. The observation that lipid bilayers are relatively permeable to water is readily explained by this mechanism.

1306

PL Yeagle

The presei~ce of cholesterol in the membrane, in particular the fused ring system of the sterol, reduces the conformational flexibility of the hydrocarbon chains of the membrane lipids, causing them to adopt an average conformation in which most carboncarbon single bonds are in the t r a n s (more ordered) configuration [11]. The increase in ordering of the lipid hydrocarbon chains leads to more effective packing of the lipids in the bilayer and a consequent reduction in the bilayer packing defects. Permeability of the membrane to small, uncharged molecules is reduced by this mechanism (see fig 4). Litman and coworkers have approached this issue from a different perspective, using the properties of a fluorescent probe to calculate the effective 'free volume" in a membrane [29]. The concept of free volume is related to the defects in packing just discussed. The addition of cholesterol reduces the 'free volume" in the membrane [30]. Figme 5 shows an example of the influence of cholesterol on this 'free volume' parameter. Also shown in figure 5 is the influence on the 'free volume' parameter of the incorporation of an integral membrane protein in the membrane bilayer [31]. The example shown is the reconstitution of rhodopsin into a phosphatidylcholine bilayer. Clearly, a membrane protein, upon incorporation into a lipid bilayer, reduces the available 'free volume', analogous to the incorporation of cholesterol into the membrane. This observation can be understood by a requirement of the membrane protein upon incorporation into a membrane for some response by the bilayer to accommodate the volume occupied by the protein. Scavenging free volume from packing defects in the lipids may provide part of the volume required by a

membrane protein undergoing normal 'breathing modes' as the protein samples a limited set of conformations within the bilayer, the extent of which is determined, in part, by the ambient temperature. One would already anticipate that the influence of cholesterol on the properties of the lipid bilayer may be antagonistic to the need for 'free volume' by the integral membrane protein. In such a case, cholesterol might be expected to inhibit the function of such a membrane protein. Sterol-protein interactions

The other interaction between cholesterol and membrane components that could be important to membrane function is a direct interaction between the sterol and the membrane protein. Studies have shown an apparent binding of cholesterol to some integral membrane proteins, including glycophorin [32] and band 3 [33, 341 from the human erythrocyte membrane. In the case of the latter protein, cholesterol has been shown to be an inhibitor of function [21]. Time-resolved fluorescence studies of a fluorescent derivative of cholesterol have also suggested cholesterol-protein interactions in the human erythrocyte membrane [35]. Furthermore, cholesterol binding to the fusion protein of the Sendai virus envelope membrane has been reported [36]. The possibility that cholesterol can bind directly to membrane proteins suggests a mechanism of sterol modulation of membrane proteins in a manner analogous to effectors of water-soluble proteins. One would predict that such a mechanism would also give 40

1.2

120 O omm

--

omm

100 N

80

¢d

20

60

40

0.8

-~

"0.6

"~'--

0.4

~

10

0 =

-~

O

$=

e~

1.0

30

O

0.2

20

0

0

0.0 0.0

0

10

20

30

40

50

60

Mole % cholesterol Fig 4. Reduction in passive glucose permeability of egg phosphatidylcholine bilayers by cholesterol. Data replotted from [ 10].

0.1

0.2

0.3

"fractional volume" Fig 5. 'Free volume' parameter (fractional volume) as determined from diphenylhexatriene fluorescence in egg phosphatidylcholine bilayers as a function of added cholesterol and rhodopsin. Data replotted from [30, 31 ].

Essential role of cholesterol in cells rise to a structural specificity in the sterol modulation of membrane proteins. The degree of such specificity would be governed by nature of the sterol-protein interactions. Interestingly, cholesterol has been suggested to not interact directly with the Ca 2+ ATPase of rabbit sarcoplasmic reticulum [37, 38]. For the same protein, cholesterol has been shown to be ineffective at modulating activity in the native membrane [39, 40], although it will stimulate activity in reconstituted membranes containing phosphatidylethanolamine which may have something to do with the effects of an interaction between cholesterol and phosphatidylethanolamine on bilayer structure [41].

A comprehensive hypothesis for cholesterol modulation of membrane function This discussion has led to the hypothesis that cholesterol likely modulates the function of biological membranes by more than one mechanism: 1) Cholesterol alters the bulk biophysical properties of membranes. Cholesterol increases the orientational order of the lipid hydrocarbon chains of membranes and reduces the 'free volume' available to membrane proteins for conformation changes that may be required for membrane protein function. In this role, cholesterol likely inhibits membrane function. 2) Cholesterol may bind directly to membrane proteins and regulate their function. In this role cholesterol may stimulate or may iqhibit membrane function. The discussion can now return to the specific cases provided above for cholesterol modulation of membrane proteins. The first case was that of bovine rhodopsin reconstituted into membranes containing varying levels of cholesterol [ 16]. Increasing cholesterol levels inhibited the Meta I to Meta II transition. Increasing cholesterol levels also inhibited the ability of this photoreceptor to activate the cGMP cascade [ 17]. Comparison of figures 1 and 4 shows that the relationships described for cholesterol effects on 'free volume' and cholesterol inhibition of the Meta I to Meta II transition are remarkably similar. These data suggest that: (1) 'free volume' may be required by the protein for function; (2) cholesterol reduces the 'free volume' available to the protein in the bilayer; (3) cholesterol thereby inhibits the function of rhodopsin in the membrane. The second case provided above for cholesterol modulation of membrane function was that of the Na+-K+-ATPase. Cholesterol stimulates the Na+K+ATPase at low to moderate membrane cholesterol levels (see fig 2). Lanosterol and ergosterol were shown to be unable to fully substitute for cholesterol in this stimulation of the enzyme [42].

1307

Figure 6 regraphs the data for several different enzymes describing the dependence of these enzymes on cholesterol level in the membrane. Each of these examples reveals a stimulation of enzyme activity at low membrane cholesterol content and an inhibition of enzyme activity at high membrane cholesterol content. Each enzyme exhibits maximal activity at somewhat different cholesterol levels in the membrane, although some of these enzymes operate near maximum activity at the level of cholesterol in the plasma membrane in which these enzymes reside. Furthermore, the data predict that these enzymes exhibit little, or no activity in the absence of cholesterol. These results suggest that cholesterol is required for these enzymes to express normal activity. More recently, an extensive study of the influence of several sterols on the activity of the Na÷-K +ATPase provided greater insight into the specificity of the stimulation for sterol structure [22]. For each of these sterols, previous measurements have provided a measure of the ability of these sterols to reduce the packing defects and favor an all-trans conformation of the carbon-carbon single bonds in the lipid hydrocarbon chains [43]. One such parameter is the reduction in the area per headgroup for the phospholipids in monolayers. This is a measure of the so-called 'condensing effect' of sterols. Figure 7 shows a plot of the extent of the 'condensing effect' of the sterols (reduction in area per lipid headgroup at a fixed sterol level in the monolayer of phospholipid) as a function of the ability of each of these sterols to support ATP hydrolyzing activity of the Na+-K+-ATPase. Each point represents a different ~l-~.,-.I ] t / i ~ n4. . . . . ,. ~,,,.,,L is evident from this gl'dpll is that no "

.

.

.

.

120 100 80 •-~

60 40 20 0 0

20

40

60

80

Mole % chol Fig 6. Replot of data on the effects of cholesterol on membrane enzymes. ([]), Na, Ca exchange [22]; (A), Na+K+-ATPase [42]; (ll), glutamate transport [51]; (A), GABA transport [51 ].

1308

PL Yeagle

simple direct proportionality exists between the ability of the sterol to "condense' the phospholipids in a monolayer and the ability of the sterols to support activity of the enzyme under otherwise constant conditions in a reconstituted system. A similar lack of a simple relationship is observed between the reduction by sterol in passive permeability of the membrane to small solutes and the activity of Na ÷Ca +-" exchange. (It should be noted that the details of these relationships can depend upon the phospholipid composition of the membrane [22]. This may be due to organization of cholesterol-rich regions in the plane of the membrane that is dependent upon phospholipid composition [44].) These data indicate that the activation of the Na +K÷-ATPase by sterol is not a process that is modulated by the condensing or ordering of the membrane lipids by cholesterol. Not only is there no correlation in figure 7 with enzyme activity, but the small amounts of cholesterol that are required to activate some of the enzymes (see figure 6) do not have an equivalently dramatic effect on the ordering of the hydrocarbon chains of the membrane lipids at those same low cholesterol levels. At the membrane cholesterol levels that do significantly affect the bulk properties of the lipid bilayer, one observes inhibition of enzyme activity (see below). 200 Qull m~

/

100 [-

[]

Z 0

• 0

L'

mr3

!

!

5

10

•

15

A area/molecule Fig 7. Relationship between the Na÷-K+-ATPase activity in the presence of various sterols reconstituted in mixed bilayers of phosphatidylcholine/phosphatidylserine/sterol (mol ratio: 30:50:20) [22] and the condensing effect of cholesterol on phosphatidylcholine bilayers, represented as the additional reduction in area per headgroup due to the presence of the sterol in the membrane at 50 mol percent [43]. The data for the latter were obtained at 22°C in phosphatidylcholine. The sterols represented in this graph are epicholesterol, ergosterol, stigrnasteroi, 7-dehydrocholesterol, cholesterol, and cholestanol. The square at the origin represents results in the absence of sterol.

What is suggested by the graph in figure 7 is that activation of the Na+-K+-ATPase by sterol at low to moderate sterol levels in the membrane is highly structurally specific. Such structural specificity can be best explained by a direct interaction of sterol with membrane protein. The shape of the activation curve of the Na+-K+-ATPase is similar to a binding isotherm. Although such direct interactions between sterol and protein have not yet been studied in these enzyme systems, the data reviewed previously in this article indicated that direct sterol-protein interactions involving membrane proteins were possible. With this model, the data on cholesterol modulation of the Na+-K+-ATPase can be explained by the competition of two effects. One is stimulation of the enzyme by cholesterol at low to moderate cholesterol levels. Stimulation in this model would result from a direct interaction of the sterol with the protein. The other is inhibition of the enzyme by cholesterol at high membrane cholesterol levels. Inhibition in this model would result from a restriction of conformational changes by the enzyme required for function due to a reduction in free volume available within the lipid bilayer by the presence of the cholesterol. Figure 2 schematically shows this division between the two effects of cholesterol on the Na÷-K+-ATPase. These two effects of cholesterol on the Na÷-K÷ATPase compete so as to produce a region of membrane cholesterol levels that support maximal activity. Interestingly for the kidney Na÷-K+-ATPase, maximal activity is observed at the cholesterol level of the native membrane [42]. This is appropriate for an organ in which the Na÷-K+-ATPase is very important to the function of that organ. Thus cholesterol homeostasis may be important to ion transport in the kidney. In contrast, in the human erythrocyte, where high fluxes of sodium and potassium are not as crucial to cellular function, the membrane cholesterol level determines operation of the erythrocyte Na+-K +ATPase at sub-maximal activity. In summa~, at this t:oint, cholesterol regulates the activity of the Na-K-ATPase in two ways. At low membrane cholesterol levels, cholesterol stimulates the enzyme. At high membrane cholesterol levels, cholesterol inhibits the Na-K-ATPase. The stimulation of the Na-K-ATPase is likely due to binding of cholesterol to the enzyme, analogous to modulators of water soluble enzymes. The stimulation is specific for the structure of the sterol. The inhibition is likely due to inhibition of the ability of the enzyme to undergo the conformational changes required for function through the effects of high membrane cholesterol on the bulk properties of the membrane. For those enzymes in which only inhibition by cholesterol is observed, one could postulate that there are no sites for effective interaction between sterol

Essential role of cholesterol in cells

12°t

of bulk effects on membrane properties yields the same conclusion as found above for the Na+-K ÷ATPase. Figure 8 shows a plot of the ability of sterols to promote cell growth versus the condensing effect of the sterols. As in the case of the Na+-K+-ATPase, no simple relationship is observed. Rather, it again appears that structurally specific interactions involving cholesterol and cellular components capable of specific recognition of the chemical structure of cholesterol, such as membrane proteins, are crucial to the growth of the cell.

100 d~ 80 O

60 iml m

40 20 012-- . . . . 0

1309

[] , 5

. . . .

[], • • 10

References

15

delta area/tool Fig 8. Relationship between the influence of sterols on cell growth of a human macrophage-like cell line [49] and the condensing effect of the sterols (see fig 7). and protein. In this case, the only effect would be the influence o f cholesterol on the bulk properties of the bilayer, thereby indirectly affecting the ability of membrane proteins to undergo conformational changes. O f course, the other possibility of cholesterol acting as a negative effector by binding to the enzyme is possible and m a y be important to regulation of the H M G C o A reductase [45, 46]. One other case should be considered: the case of the membrane protein whose activity is not affected by the presence of cholesterol in the membrane. The Ca 2+ ATPase in the rabbit muscle sarcoplasmic reticulum provides an example (see above). In this case there would presumably be no sites on the protein at which cholesterol could bind. Available data suggest that this is the case [47]. Furthermore, the portion of the protein in the membrane m a y not have to undergo significant conformational changes that absorb 'free v o l u m e ' from the lipid bilayer during its functional cycle. That m a y be reasonable for this protein in which the ATPase active site is on the portion of the protein that is outside of the lipid bilayer [48]. Based on the above analysis, it is possible to formulate an hypothesis for the essential role of cholesterol in cholesterol-requiring cells: that the essential role of cholesterol in cell function is to activate membrane enzymes that are necessary for cellular function and growth. A mechanism for such structurally specific sterol stimulation of membrane function would be interactions between cholesterol and membrane proteins that are structurally specific. Studies on the sterol requirement of a human macrophage-like cell line support this hypothesis in vivo. A highly specific sterol requirement was found for growth of this cell line in sterol-depleted media [49]. Examining this sterol requirement for evidence

1 Faust JR, Trzaskos JM, Gaylor JL (1988) Cholesterol Biosynthesis. In: Biology of Cholesterol (Yeagle PL, ed) CRC Press, Boca Raton, FL 2 Bloch K (1976) On the evolution of a biosynthetic pathway. In: Reflections in Biochemistry (Komberg A, Korecker BLH, Comudella Z, Oro J, eds) Pergamon Press, Oxford 3 Dahl C, Dahl J (1988) Cholesterol and cell function. In: Biology of Cholesterol (Yeagle PL, ed) CRC Press, Boca Raton, FL 4 Dahl JS, Dahl CE, Bloch K (1980) Sterol in membranes: Growth characteristics and membrane properties of Mycoplasma capricolum cultured on cholesterol and lanosterol. Biochemistry 19, 1467-1472 5 Andreason AA, Stier TJB (1953) Anaerobic nutrition of Saccharomyces cerevisiae. Ergosterol requirements for growth in a defined medium. J Cell Comp Physiol 41, 23 6 Chen HW, Kandutsch AA, Waymouth C (1974) Inhibition of cell growth by oxygenated derivatives of cholesterol. Nature 251, 419 7 Brown MS, Goldstein JL (1974) Suppression of 3hydroxy-3-methylglytaryl-coenzyme A reductase activity and inhibition of growth of human fibroblasts of 7-ketocholesterol. J Biol Chem 249, 7306 8 Yeagle PL (1989) The Biology of Cholesterol. CRC Press, Boca Raton, FL 9 Huang CH (1976) Roles of carbonyl oxygens at the bilayer interface in phospholipid-sterol interaction. Nature 259, 242-244 10 Demel RA, Bruckdorfer KR, Deenen LLMV (1972) The effect of sterol structure on the permeability of liposomes to glucose, glycerol and Rb÷. Biochim Biophys Acta 255, 321-330 11 Gaily HU, Seelig A, See!ig J (1976) Cholesterol-induced rod-like motion of fatty acyl chains in lipid bilayers, a deuterium magnetic resonance study. Hoppe-Seyler's Z Physiol Chem 357, 1447-1450 12 Stockton BW, Smith ICP (1976) A deuterium NMR study of the condensing effect of cholesterol on egg phosphatidylcholine bilayer membranes. Chem Phys Lipids 17, 251 13 Smutzer G, Yeagle PL (1985) Phase behavior of DMPCcholesterol mixtures; a fluorescence anisotrophy study. Biochim Biophys Acta 814, 274-280 14 Estep TN, Mountcastle DB, Biltonen RL, Thompson TE (1978) Studies on the anomalous thermotropic behavior of aqueous dispersions of dipalmitoylphosphatidylcholinecholesterol mixtures. Biochemistry 17, 1984-1989 15 Yeagle PL (1985) Cholesterol and the cell membrane. Biochim Biophys Acta 822, 267-287

13 i 0 16

17 18

19

20

21 22 23 24

25

26 27

PL Yeagle

Mitchell DC, Straume M, Miller JL, Litman BJ (1990) Modulation of metarhodopsin formation by cholesterolinduced ordering of bilayer lipids. Biochemistry 29, 9143 Boesze-Battaglia K, Albert A (1990) Cholesterol modulation of photoreceptor function in bovine rod outer se~mrnents. J Biol Chem 265, 20727-20730 Brasitus TA, Dahiya R, Dudeja PK, Bissonnette BM (1988) Cholesterol modulates alkaline phosphatase activity of rat intestinal microvillus membranes. J Biol Chem 263, 8592-8597 Rotenbe~ M, Zakim D {1991) Effects ~f cholesterol on t,-~e function and thermo~'opic properties of pure UDP-glucuronosyltransferase. J Biol Chem 266, 41594161 Saito Y, Silbert DF (1979) Selective effects of membrane sterol depletion on surface function thymidine and 3-0methyl-D-glucose transport in a sterol auxotroph. J Biol Chem 255, 1102-1107 Gregg VA, Reithmeier RAF (1983) Effect of cholesterol on phosphate uptake by human red blood cells. FEBS Lett 157, 159--164 Vemuri R, Philipson KD (1989) Influence of sterols and phospholipids on sarcol;mmal and sarcoplasmic reticular cation transporters. J Biol Chem 264, 8680--8685 Kramer R (1982) Cholesterol as activator of ADP-ATP exchange in reconstituted liposomes and in mitochondria. Biochim Biophys Acta 693, 296-304 Grunze M, Forst B, Deuticke B (1980) Duel effect of membrane cholesterol on simple and mediated transport process in human erythrocytes. Biochim Biophys Acta 600, 860-868 Craido M, Eibl H, Barrantes FJ (1982) Effects of lipids on acetylcholine receptor. Essential need of cholesterol for maintenance of agonist-induced state transitions in lipid vesicles. Biochemistry. 21, 3622-3627 Fong TM, McNamee MG (1986) Correlation between acetylcholine receptor function and structural properties of

Yeagle PL (1987) The Membranes of Cells. Academic Press, Orlando, FL 28 Seelig A, Seelig J (1974) The dynamic structure of fatty acyl chains in a phospholipid bilayer measured by deuterium magnetic resonance. Biochemistry 13, 48394845 29 Straume M, Litman BJ (1987) Equilibrium and dynamic structure of large, unilamellar, unsaturated acyl chain phosphatidylcholine vesicles. Higher order analysis of 1,6-diphenyl-l,3,5-hexatriene and 1-[4-trimethylammonio]-6-phenyl-l,3,5-hexatriene anisotropy decay. Biochemistry. 26, 5113-5120 30 Straume M, Litman BJ (1987) Influence of cholesterol on equilibrium and dynamic bilayer structure of unsaturated acyl chain phosphatidylcholine vesicles as determined from higher order analysis of fluorescence anisotropy decay. Biochemistry 26, 5121-5 ! 26 31 Straume M, Litman BJ (1988) Equilibrium and dynamic bilayer structural properties of unsaturated acyl chain phosphatidylcholine-cholesterol-rhodopsin recombinant vesicles and rod outer segment disk membranes as determined from higher order analysis of fluorescence anisotropy decay. Biochemistry 27, 7723-7733 32 Yeagle PL (1984) Incorporation of the human erythrocyte sialoglycoprotein into recombined membranes contair.,'ng cholesterol. J Membr Bio178, 201-210

33 34 35 36

37

38 39

40

41

42 43 44 45

46

47 48

49 50 51

Klappauf E, Schubert D (1977) Band 3 from human erythrocyte membranes strongly interacts with cholesterol. FEBS Lett 80, 423-425 Schubert D, Boss K (1982) Band 3 protein-cholesterol interactions in erythrocyte membranes. FEBS Lett 150, 4-8 Yeagle PL, Albert AD, Boesze-Battaglia K, Young J, Frye J (1990) Cholesterol dynamics in membranes. Biophys J 57, in press Asano K, Asano A (1988) Binding of cholesterol and inhibitory peptide derivatives with the fusogenic hydrophobic sequence of F-glycoprotein of Sendai virus: possible implication in the fusion reaction. Biochemistry 27, 1321-1329 Warren GB, Houslay MD, Metcalfe JC, Birdsall NJM (1975) Cholesterol is excluded from the phospholipid annulus surrounding an active calcium transport protein. Nature 255,684-687 London E, Feigenson GW (1978) Fluorescence quenching of Ca2+-ATPase in bilayer vesicles by a spin-labeled phospholipid. FEBS Lett 96, 51-54 Johannsson A, Keightley CA, Smith GA, Metcalfe JC (1981) Cholesterol in sarcoplasmic reticulum and the physiological significance of membrane fluidity. Biochem J 196, 505-511 Madden TD, King MD, Quinn PJ (1981) The modulation of Ca2+-ATPase activity of sarcoplasmic reticulum by membrane cholesterol. The effect of enzyme coupling. Biochim Biophys Acta 641,265-269 Cheng K, Lepock JR, Hui SW, Yeagle PL (1986) The role of cholesterol in the activity of reconstituted Ca ATPase vesicles containing unsaturated phosphatidylethanolamine. J Biol Chem 261,5081-5087 Yeagle PL, Rice D, Young J (1988) Cholesterol effects on bovine kidney Na+ K+ATPase hydrolyzing activity. Biochemistry 27, 6449-6452 Demel RA, Bruckdorfer KR, Deenen LLMV (1972) Biochim Biophys Acta 255, 311-320 Hui SW (1988) The spatial distribution of cholesterol in membranes. In: Biology of Cholesterol (Yeagle PL, ed) CRC Press, Boca Raton, FL Heusden GHPV, Wirtz KWA (1984) Hydroxylmethylglutaryl CoA reductase and the modulation of microsomal cholesterol content by the nonspecific lipid transfer protein. J Lipid Res 25, 27-31 Bjorkhem I , Buchmann MS, Skrede S (1985) On the structural specificity in the regulation of the hydroxymethylglutaryl- CoA reductase and the cholesterol-7a-hydroxylase in rats. Effects of cholestanol feeding. Biochim Biophys Acta 835, 18-22 Warren GB, Houslay MD, Metcalfe JC, Birdsall NJM (1975) Cholesterol is excluded from the annulus of the calcium ATPase. Nature 255, 684-687 MacLennan DH, Brandl C J, Korczak B, Green NM (1985) Amino-acid sequence of a Ca 2÷ + Mg2+-dependent ATPase from rabbit muscle sarcoplasmic reticulum, deduced from it~ complementary DNA sequence. Nature 316, 696-700 Esfahani M, Scerbo L, Devlin TM (1984) A requirement for cholesterol and its structural features for a human macrophage-like cell line. J Cell Biochem 25, 87-97 Selinsky BS (1984) PhD Thesis Shouffani A, Kanner BI (1990) Cholesterol is required for reconstitution of the sodium- and chloride-coupled GABA transporter from rat brain. J Biol Chem 265, 6002-6008