[13 Alterations in Membrane Cholesterol That Affect Structure and Function of Caveolae

By Eric J. Smart and Richard G. W. Anderson


A widely acknowledged function of cellular membrane is to promote biological reactions by creating a two-dimensional space that facilitates interactions among resident molecules. Less well appreciated is that cell membranes are subdivided into specific domains that compartmentalize a variety of essential activities. Three examples of membrane domains are clathrin-coated pits, focal adhesion sites, and caveolae. Each requires specific cellular machinery for their existence and is constructed from specific membrane proteins and lipids. In the case of caveolae, cholesterol is an essential lipid. Several studies have demonstrated that it functions as a structural molecule for this domain. In this chapter we describe several methods for modifying the cholesterol composition of caveolae and discuss how these changes alter the structure and function of this membrane domain.

Caveolae were originally described as small, noncoated plasma membrane invaginations.1 There is now clear evidence, however, that caveolae can be either invaginated,2 flat within in the plane of the membrane,2'3 or detached vesicles.4,5 In 1992, Rothberg et al.2 demonstrated that a 22-kDa protein called caveolin-1 was associated with a filamentous coat that decorates the cytoplasmic side of morphologically identifiable caveolae. Caveolin-1 has since been shown to be a key player in the structure and function of caveolae. Although caveolin-1 is not exclusively localized to caveolae, it is considered by many to be the defining marker for invaginated caveolae. Many but not all cells express caveolin. Interestingly, cells not expressing caveolin-1 contain plasma membrane domains that have biochemical properties similar to those of caveolae.6,7 The relationship between caveolin-containing membranes and caveolin-deficient caveola-like membrane domains remains to be determined. For purposes of this chapter, caveolae are defined

2 K. G. Rothberg, J. E. Heuser, W. C. Donzell, Y. Ying, J. R. Glenney, and R. G. W. Anderson, Cell 68,673 (1992).

3 E. J. Smart, D. C. Foster, Y. S. Ying, B. A. Kamen, and R. G. Anderson, J. Cell Biol. 124,307 (1994).

4 J. R. Henley, E. W. Krueger, B. J. Oswald, and M. A. McNiven, J. Cell Biol. 141, 85 (1998).

5 P. Oh, D. P. Mcintosh, and J. E. Schnitzer, J. Cell Biol. 141, 101 (1998).

6 R. G. W. Anderson, Annu. Rev. Biochem. 67, 199 (1998).

7 E. J. Smart, G. A. Graf, M. A. McNiven, W. C. Sessa, J. A. Engelman, P. E. Scherer, T. Okamoto, and M. P. Lisanti, Mol. Cell. Biol. 19, 7289 (1999).

as cholesterol/sphingolipid-rich membrane domains that contain caveolin-1 and are capable of internalizing molecules.

Caveolae, Caveolin, and Cholesterol

The morphological and functional behavior of caveolae depends on specific lipids and proteins. A key molecule is cholesterol. Indeed, most of the current methods for modulating caveolar function depend on perturbing the cholesterol composition of this domain. Examples of how cholesterol affects caveolar structure and function are outlined below. Specific methods for modulating cholesterol in caveolae are described in the following sections.

Invaginated caveolae are easily seen in thin-section electron microscopy (EM) images of cells. Rapid-freeze deep-etch EM has shown that these invaginated caveolae are decorated with a filamentous coat structure.2 These images also provided strong evidence that caveolar shape is variable, ranging from flat to deeply invaginated. The first indication that cholesterol was an essential molecule of caveolae came from the discovery that depletion of membrane cholesterol inhibits the clustering of glycosylphosphatidylinositol (GPI)-anchored membrane proteins in caveolae.8,9 These studies also showed that cholesterol depletion causes the loss of invaginated caveolae. Subsequent studies demonstrated that pharmacological reagents (see below) that bind cholesterol flatten caveolae and cause disassembly of the filamentous coat structure. Biochemical assays have confirmed that caveolae from control cells have a 5-fold higher cholesterol-to-protein ratio than caveolae from cholesterol-depleted cells.10

Cholesterol depletion also affects caveolar function. Caveolae in the kidney epithelial cell line MA 104 are involved in the selective uptake of 5-methyltetra-hydrofolic acid.11,12 The uptake is selective because 5-methyltetrahydrofolic acid is delivered to the cytoplasm but the receptor remains associated with the cell membrane. This process is called potocytosis.12 Internalization involves the binding of the vitamin to the folate receptor, which is a glycosylphosphatidylinositol (GPI)-anchored receptor enriched in caveolae.9,11'13 In these cells, caveolae cycle between an open or extracellular exposed conformation and a closed or sequestered conformation once every 60 min. When caveolae close, a bafilomycin-sensitive proton pump transports protons into the caveolae lumen, thereby decreasing the

8 W. J. Chang, K. G. Rothberg, B. A. Kamen, and R. G. Anderson, J. Cell Biol. 118, 63 (1992).

9 K. G. Rothberg, Y. Ying, B. A. Kamen, and R. G. W. Anderson, J. Cell Biol. Ill, 2931 (1990).

10 E. J. Smart, Y. S. Ying, P. A. Conrad, and R. G. Anderson, J. Cell Biol. 127, 1185 (1994).

11 K. G. Rothberg, Y. Ying, J. F. Kolhouse, B. A. Kamen, and R. G. W. Anderson, J. Cell Biol. 110, 637(1990).

12 R. G. Anderson, B. A. Kamen, K. G. Rothberg, and S. W. Lacey, Science 255,410 (1992).

13 W.-J. Chang, K. G. Rothberg, B. A. Kamen, and R. G. W. Anderson, J. Cell Biol. 118, 63 (1992).

pH. The ligand dissociates from the receptor and translocates to the cytosol through an anion transporter. The caveolae then reopen and the receptors are free to repeat the cycle. A critical feature of this mechanism is the ability of caveolae to completely invaginate and form a sequestered compartment, a process that may involve pinching off from the membrane to form a vesicle. The potocytosis of 5-methyltetrahydrofolic acid is inhibited when caveolae are depleted of cholesterol but resumes on sterol repletion.9'13

In addition to being the vehicle for potocytosis, caveolae also compartmentalize a variety of signaling activities that take place at the cell surface.6'7'14 Caveolar cholesterol is essential for signal transduction that originates in this domain. Many signaling molecules enriched in caveolae are either acylated or prenylated and cholesterol depletion causes their mislocalization.6'714 This leads to an alteration in signal transduction by these molecules. For example, the localization of endothelial nitric-oxide synthase (eNOS) to caveolae requires that it be both myristoylated and palmitoylated.15 Depletion of caveolae cholesterol in endothelial cells, using either oxidized low-density lipoprotein or cyclodextrin (see below), causes the relocalization of eNOS to an intracellular compartment without a corresponding alteration in the acylation state of the enzyme.16'17 Relocated eNOS no longer is activated by physiological stimuli such as acetylcholine.16'17 Cholesterol is also required for the appropriate activity of many signaling molecules that reside in caveolae. Depletion of caveolar cholesterol in fibroblasts, for example, causes spontaneous activation of Extracellular signal-regulated kinase-1/2 (ERK1/2) and the addition of epithelial growth factor (EGF) causes further activation of this crucial regulatory kinase.18

Caveolin-1 is thought to play a role in maintaining the proper cholesterol level of caveolae. Exactly how it does this is not clear. The caveolae in cells that express caveolin-1 have a higher cholesterol-to-protein ratio than the surrounding membrane. A cytosolic pool of caveolin-1 in a complex with several heat shock proteins is an intermediate in transport of cholesterol from the endoplasmic reticulum (ER) to caveolae.19 Palmitoylation of caveolin-1 appears to be necessary to promote association of cholesterol with caveolin-1 and the formation of cholesterol-rich caveolae.20 Caveolin-1 has also been shown to function

14 P. W. Shaul and R. G. Anderson, Am. J. Physiol. 275, L843 (1998).

15 P. W. Shaul, E. J. Smart, L. J. Robinson, Z. German, I. S. Yuhanna, Y. Ying, R. G. Anderson, and T. Michel, J. Biol. Chem. 271, 6518 (1996).

16 A. Uittenbogaard, P. W. Shaul, I. S. Yuhanna, A. Blair, and E. J. Smart, J. Biol. Chem. 275, 11278 (2000).

17 A. Blair, P. W. Shaul, I. S. Yuhanna, P. A. Conrad, and E. J. Smart, J. Biol. Chem. 274,32512 (1999).

18 T. Furuchi and R. G. Anderson, J. Biol. Chem. 273, 21099 (1998).

19 A. Uittenbogaard, Y. Ying, and E. J. Smart, J. Biol. Chem. 273, 6525 (1998).

20 A. Uittenbogaard and E. J. Smart, J. Biol. Chem. 275, 25595 (2000).

as an apolipoprotein in certain cells and to be secreted in lipid particles that have the characteristics of high-density lipoprotein (HDL).21 Caveolin-1 may also function as a scaffolding protein for multiple signaling molecules. It contains a short amino acid sequence that has been termed the scaffolding domain.22 Numerous in vitro binding studies have shown that this region of the molecule interacts with proteins that have a characteristic amino acid sequence called the caveolin-1-binding domain.22-25 Thus, the ability of caveolin-1 to form a coat structure may position the scaffolding domain so that it directly binds to signaling molecules containing the caveolin-1-binding motif and maintain the protein in an inactive state.7,14 Many methods that alter caveolar cholesterol levels cause caveolin-1 to move to other cellular locations. Regardless of its function, however, depletion of caveolin-1 from caveolae will perturb signal transduction from this domain.

Cholesterol-Binding Drugs

The first pharmacological reagents identified that disrupts the structure of caveolae were cholesterol-binding drugs such as filipin and nystatin.9 Filipin is a polyene macrolide antibiotic that binds to cholesterol and disrupts the organization of the surrounding membrane.26,27 de Kruijff and Demel proposed that plasma membrane cholesterol associates with filipin to generate bulky complexes that deform the plasma membrane.28 Despite the possibility of nonspecific effects, filipin will disrupt the caveolar coat structure, cause caveolae to flatten, and prevent internalization of caveolae.9 Another important feature of filipin is that it does not inhibit clathrin-mediated endocytosis, which makes it potentially useful for distinguishing between these two endocytic pathways.9


1. Prepare a fresh 10-mg/ml stock solution of filipin in dimethyl sulfoxide (DMSO) (filipin complex is from Sigma, St. Louis, MO).

2. Wash the cells once with phosphate-buffered saline (PBS).

3. Add fresh culture medium without serum to the cells.

21 P. Liu, W. P. Li, T. Machleidt, and R. G. Anderson, Nat. Cell Biol. 1, 369 (1999).

22 S. Li, J. Couet, and M. P. Lisanti, J. Biol. Chem. 271,29182 (1996).

23 J. Couet, S. Li, T. Okamoto, T. Ikezu, and M. P. Lisanti, J. Biol. Chem. 272, 6525 (1997).

24 J. A. Engelman, C. Chu, A. Lin, H. Jo, T. Ikezu, T. Okamoto, D. S. Kohtz, and M. P. Lisanti, FEBS Lett. 428, 205 (1998).

25 G. Garcia-Cardena, P. Martasek, B. S. Masters, P. M. Skidd, J. Couet, S. Li, M. P. Lisanti, and W. C. Sessa, J. Biol. Chem. 272, 25437 (1997).

26 J. Bolard, Biochim. Biophys. Acta 864, 257 (1986).

27 J. Milhaud, Biochim. Biophys. Acta 1105, 307 (1992).

28 B. de Kruijff and R. A. Demel, Biochim. Biophys. Acta 26, 57 (1974).

4. Add filipin to achieve a final concentration of 5-10 /xg/ml [DMSO <0.05% (v/v)].

5. Incubate the cells at 37° for 15-60 min.

Potential Problems. The ability of filipin to permeabilize membranes makes it necessary to determine the concentration that will affect caveolar function without damaging the cell. At low concentrations, filipin will disrupt caveolar function with minimum side effects. Too high a concentration, however, will dramatically affect cell morphology, cell permeability, and cell viability. Trypan Blue exclusion, lactate dehydrogenase release, and transferrin uptake are standard tests of cell viability that should be used to establish how each cell type responds to the drug. As with any reagent, filipin can have side effects and should not be considered a specific inhibitor of caveolar function. Nevertheless, when combined with other tests, filipin can be useful for identifying events in the cell that depend on caveolae.

Cholesterol Oxidase

Cholesterol oxidase is a bacterial enzyme that converts cholesterol to cholest-4-en-3-one.29 The activity of the enzyme on membrane cholesterol is greatly influenced by the local lipid environment within the membrane. The lateral surface pressure of the membrane, the phospholipid composition, and the amount of cholesterol are all critical factors that influence the function of cholesterol oxidase.29,30 One of the first uses of cholesterol oxidase was for detecting the distribution of cellular cholesterol in glutaraldehyde-fixed cells.30 Nevertheless, cholesterol oxidase will work on live cells, where it preferentially oxidizes caveolar cholesterol and causes caveolin-1 to accumulate in the Golgi apparatus.10 This property of cholesterol oxidase has found use in studying signal transduction from caveolae as well as the transport of cholesterol to caveolae.


1. Cholesterol oxidase is from Boehringer Mannheim (Indianapolis, IN) (Rhodococcus erythropolis, 25 units/mg).

2. Wash cells once with PBS.

3. Add fresh cell culture medium without serum to the cells.

4. Add cholesterol oxidase to achieve a final concentration of 0.5 units/ml.

5. Incubate the cells at 37° for 60 min.

29 J. MacLachlan, A. T. L. Wotherspoon, R. O. Ansell, and C. J. W. Brooks, J. Steroid Biochem. Mol.

Potential Problems. The main difficulty with this method is that not all cells respond to cholesterol oxidase.29,30 Therefore, the activity of the enzyme needs to be tested on each cell type by assaying for the conversion of caveolar cholesterol to cholest-4-en-3-one. Cells that do respond need to be grown and handled according to a standard format. For example, the cell density can affect the ability of cholesterol oxidase to access the sterol. The presence of other cholesterol sources such as serum lipoproteins also dramatically affects the activity of the enzyme. Serum-free conditions should always be used. The temperature of the reaction is another variable that needs to be controlled. The enzyme is less active at reduced temperatures. Thus, cholesterol oxidase is an effective tool for studying the function of caveolae and caveolin-1 when the conditions are optimized and the activity of the enzyme is determined for each experiment.


Cyclodextrins are gaining wide acceptance as a tool for modifying the structure and function of caveolae. /^-Cyclodextrins are cyclic heptasaccharides consisting of /}(l-4)-glucopyranose units.31 Numerous cyclodextrins exist but the one most effective at removing membrane cholesterol is methyl-/3-cyclodextrin.32 This cyclodextrin is water soluble and contains a hydrophobic core that has a specific affinity for cholesterol. Cholesterol bound to methyl-/3-cyclodextrin is soluble but can exchange with plasma membrane cholesterol. Moreover, the addition of methyl-yS-cyclodextrin to cells effectively removes cholesterol from living cells. The efficiency of extraction varies with cell type.33 Cyclodextrins remove cholesterol from all parts of the membrane, not just caveolae. Work by Haynes et al. suggests that for each cell type conditions can be found wherein methyl-/}-cyclodextrin will selectively extract cholesterol from different membrane pools of the sterol.34 Selective removal of cholesterol from caveolae may be favored at low concentrations of methyl-/)-cyclodextrin because of the high cholesterol content of this membrane domain.


1. Prepare a fresh 500 mM stock solution of methyl-/?-cyclodextrin in water.

2. Wash cells once with PBS.

3. Add fresh cell culture medium without serum to the cells.

4. Add methyl-/3-cyclodextrin to a final concentration of 5 mM.

31 J. Pitha, T. Irie, P. B. Sklar, and J. S. Nye, Life Sci. 43,493 (1988).

32 T. Irie, K. Fukunaga, and J. Pitha, J. Pharm. Sci. 81, 521 (1992).

33 E. P. Kilsdonk, P. G. Yancey, G. W. Stoudt, F. W. Bangerter, W. J. Johnson, M. C. Phillips, and G. H. Rothblat, J. Biol. Chem. 270, 17250 (1995).

34 M. P. Haynes, M. C. Phillips, and G. H. Rothblat, Biochemistry 39,4508 (2000).

Potential Problems. There are several factors that influence the ability of methyl-/}-cyclodextrin to remove cholesterol. Because methyl-/i-cyclodextrin can promote both efflux and influx of cholesterol, the culture medium cannot contain a source of cholesterol such as serum. Long-term incubation of cells in the presence of methyl-/}-cyclodextrin (>6 hr) will reduce cellular cholesterol and upregulate genes that control endogenous cholesterol levels. It can also affect cell viability. Each experimental paradigm needs to be tested to ensure that methyl-/?-cyclodextrin is removing caveolar cholesterol without causing massive changes in the total plasma membrane cholesterol. Therefore, tests should be performed to determine the amount of cholesterol in the caveolar fraction compared with the total plasma membrane after methyl-/}-cyclodextrin treatment. Finally, if methyl-/3-cyclodextrin is observed to have an affect, the addition of cholesterol to the medium should block this affect.

Oxidized Low-Density Lipoprotein

Oxidized low-density lipoprotein (LDL) also alters the level of caveolar cholesterol and the localization of caveolin-1 in a receptor-dependent manner.16,17 Oxidized LDL bound to CD36 receptors located in caveolae promotes the efflux of caveolar cholesterol. In addition, we have demonstrated that HDL, in a scavenger receptor class B, type I (SR-BI)-dependent manner, delivers sterol to caveolae and counteracts the effects of oxidized LDL.16,17 Cells must express CD36 in caveolae for this method to work. Immunoblotting of isolated caveolar fractions and immunofluorescence are the recommended methods for determining whether CD36 is localized to caveolae. The preparation of LDL and oxidized LDL has been described elsewhere.35"37


1. Prepare fresh oxidized LDL [5-15 TBARs (thiobarbituric acid-reactive substances)] as described Buege and Aust.37

2. Wash cells once with PBS.

3. Add fresh cell culture medium without serum to the cells.

4. Add oxidized LDL to a final concentration of 10-50 /ig/ml.

Potential Problems. Cells should not be exposed to oxidized LDL for more than a few hours. Lipoproteins potentially can affect the level of caveolin-1 in cells.

35 W. R. Fisher and V. N. Schumaker, Methods Enzymol. 128, 247 (1986).

36 J. L. Kelley and A. W. Kruski, Methods Enzymol. 128, 170 (1986).

37 J. A. Buege and S. D. Aust, Methods Enzymol. 52, 302 (1978).

In addition, LDL bound to CD36 may activate peroxisome proliferator-activated receptor y (PPARy)-regulated genes. As with all the experimental protocols, the state of cholesterol in the caveolar fraction needs to be determined.

Quantification of Cholesterol

Because it is necessary to determine the amount of cholesterol associated with caveolae and plasma membranes, we have included two common methods for quantifying cholesterol: radioisotope labeling and direct mass measurement. Both these methods are reliable and widely used. The radioisotope labeling method is useful for pulse-chase experiments and distinguishing newly synthesized cholesterol from existing cholesterol pools. The direct mass measurement method circumvents potential problems with exchange and isotope dilution but is less sensitive than isotope labeling.

Method: Radioisotope Labeling

See Smart et al.3S for details.

1. Wash cells once with an acetate-free buffer such as PBS.

2. Add medium and lipoprotein-deficient serum to the cells.

3. Add unlabeled acetate to a final concentration of 10 ¡iM and add 500 /zCi of [3H]acetate (New England Nuclear, Boston, MA) to one 150-mm plate of cells.

5. Process the cells as required (fractionation, lysate, etc.).

6. Adjust the volume of each sample to 1 ml with water and transfer the sample to a 13 x 100 mm glass tube.

7. Add 1.2 ml of 2% (v/v) acetic acid in methanol and mix thoroughly.

8. Add 1.2 ml of chloroform and mix thoroughly.

9. Centrifuge at lOOOg for 15 min at room temperature.

10. Take the bottom layer (chloroform) and transfer it to a microcentrifuge tube.

11. Add 5 /ii of cholesterol (10 mg/ml in chloroform) to each sample to aid in the identification of the appropriate spots on the thin-layer chromatography plates.

12. Dry the samples under nitrogen.

13. Suspend samples in the thin-layer chromatography solvent system [petroleum ether-ethyl ether-acetic acid, 80:20:1 (v/v/v)].

14. Spot the samples and standards onto a silica gel plate [Si250-PA (19C); VWR, Scientific, San Francisco, CA].

15. Place the plate in the solvent chamber and allow the solvent to migrate to within 1 in. of the top of the plate.

38 E. J. Smart, Y.-S. Ying, W. C. Donzell, and R. G. Anderson, J. Biol. Chem. 271, 29427 (1996).

16. Remove the plate and air dry.

17. Lightly spray the plate with a fresh solution of 5% (v/v) sulfuric acid made in ethanol.

18. Let the plate completely air dry and then bake for 10 min at 170°.

19. Scrape the appropriate spots into scintillation fluid and count.

Method: Direct Mass Determination See Uittenbogaard et al.16 for details.

1. Process the cells as required (fractionation, lysate, etc.).

2. Adjust the volume of each sample to 1 ml with water and transfer the sample to a 13 x 100 mm glass tube.

3. Add 3 ml of chloroform-acetic acid (1:2, v/v) and mix.

4. Add 1 ml of chloroform and mix.

6. Centrifuge at lOOOg of 10 min at room temperature.

7. Take the bottom layer (chloroform) and transfer to a 10 x 75 mm glass tube.

8. Completely dry the samples and standards [1, 2, 3, 5, 10, and 20 fig of standard solution from a Wako (Tokyo, Japan) cholesterol determination kit] with nitrogen.

9. Add 1 ml of 1% (v/v) Triton X-100 made in chloroform to each sample.

10. Dry with nitrogen. The sample will coat the tube.

11. Add 0.5 ml of water to dissolve the Triton X-100 and sample.

12. Add 0.5 ml of 2x enzyme mix to samples and standards. The enzyme mix is from a commercial cholesterol determination kit (Wako). The enzyme mix must be diluted fresh each time.

13. Incubate the samples and standards at 37° for 15 min.

14. Measure the absorbance at 505 nm.


Most of the available methods for modifying caveolae structure and function depend on altering the cholesterol content of caveolae. The most important aspect of each method is to ensure the reagents are working in the cells that are being studied. The idiosyncrasies of each method are such that they cannot be universally applied without carefully optimizing the conditions. When used correctly, these methods are accepted as a specific way to perturb the structure and function of caveolae.

Lower Your Cholesterol In Just 33 Days

Lower Your Cholesterol In Just 33 Days

Discover secrets, myths, truths, lies and strategies for dealing effectively with cholesterol, now and forever! Uncover techniques, remedies and alternative for lowering your cholesterol quickly and significantly in just ONE MONTH! Find insights into the screenings, meanings and numbers involved in lowering cholesterol and the implications, consideration it has for your lifestyle and future!

Get My Free Ebook

Post a comment