Traffic 2010; 11: 348–360 © 2009 John Wiley & Sons A/S

doi:10.1111/j.1600-0854.2009.01022.x

Gangliosides and β1-Integrin Are Required for Caveolaeand Membrane Domains

Raman Deep Singh1, David L. Marks1, Eileen L.

Holicky1, Christine L. Wheatley1, Tatiana

Kaptzan1, Satoshi B. Sato2, Toshihide

Kobayashi3, Kun Ling1 and Richard E. Pagano1,∗

1Departments of Medicine, Biochemistry and MolecularBiology, Mayo Clinic College of Medicine, Rochester, MN55905, USA2Department of Biophysics, Graduate School of Science,Kyoto University, Kyoto 606-8502, Japan3RIKEN, Wako, Saitama 351-0198, Japan*Corresponding author: Richard E. Pagano,pagano.richard@mayo.edu

Caveolae are plasma membrane domains involved in

the uptake of certain pathogens and toxins. Internaliza-

tion of some cell surface integrins occurs via caveolae

suggesting caveolae may play a crucial role in mod-

ulating integrin-mediated adhesion and cell migration.

Here we demonstrate a critical role for gangliosides

(sialo-glycosphingolipids) in regulating caveolar endo-

cytosis in human skin fibroblasts. Pretreatment of cells

with endoglycoceramidase (cleaves glycosphingolipids)

or sialidase (modifies cell surface gangliosides and glyco-

proteins) selectively inhibited caveolar endocytosis by

>70%, inhibited the formation of plasma membrane

domains enriched in sphingolipids and cholesterol (‘lipid

rafts’), reduced caveolae and caveolin-1 at the plasma

membrane by approximately 80%, and blunted activation

of β1-integrin, a protein required for caveolar endocyto-

sis in these cells. These effects could be reversed by a

brief incubation with gangliosides (but not with asialo-

gangliosides or other sphingolipids) at 10◦C, suggesting

that sialo-lipids are critical in supporting caveolar endo-

cytosis. Endoglycoceramidase treatment also caused a

redistribution of focal adhesion kinase, paxillin, talin, and

PIP Kinase Iγ away from focal adhesions. The effects of

sialidase or endoglycoceramidase on membrane domains

and the distribution of caveolin-1 could be recapitulated

by β1-integrin knockdown. These results suggest that

both gangliosides and β1-integrin are required for main-

tenance of caveolae and plasma membrane domains.

Key words: caveolar endocytosis, caveolin-1, endogly-

coceramidase, focal adhesions, glycosphingolipids, siali-

dase

Received 2 November 2009, revised and accepted for

publication 1 December 2009, uncorrected manuscript

published online 3 December 2009, published online 31

December 2009

Many endocytic entry pathways into cells have beenidentified. These vary in the cargo molecules they

transport and the underlying protein machinery thatfacilitates the different endocytic processes (1–3). Amongthe clathrin-independent mechanisms of endocytosis,caveolar uptake is perhaps the best studied. Caveolaeare 50–80 nm diameter flask-shaped plasma membrane(PM) invaginations that are marked by the presence ofa member of the caveolin (Cav) protein family (4) and bypolymerase I and transcript release factor- (PTRF)Cavin, aputative caveolar coat protein that is thought to be requiredfor caveola formation (5,6). Markers used to visualizeuptake through caveolae include labeled albumin (7–9),SV40 virus (10), and in some cell types, the choleratoxin B (CtxB) subunit (8,11,12). In addition, we andothers have provided evidence that a fluorescent analogof lactosylceramide (LacCer) [N-(4,4-difluoro-5,7-dimethyl-4-bora-3a,4a-diaza-s-indacene-3-pentanoyl)sphingosyl 1-β-D-lactoside; Bodipy-LacCer], and other glycosphingolipid(GSL) analogs, are internalized almost exclusively viacaveolae in human skin fibroblasts (HSFs) and other celltypes based on multiple approaches (8,9,13–15).

Although the importance of caveolar endocytosis is nowwell appreciated, the regulation of this process is notfully understood. Several factors however are known toplay an important role in this process. First, caveolarendocytosis is stimulated when cells are treated withphosphatase inhibitors (e.g. okadaic acid) or when certaincaveolar cargo binds to its receptor (16,17). Stimulationof caveolar endocytosis is accompanied by increasedactivation of src and phosphorylation of caveolin-1 (Cav1)and dynamin (18), suggesting that stimulation occurs viaincreased kinase activities. In agreement with this notion,a screen of the human kinome identified a total of 80different kinases that are somehow involved in the uptakeof SV40 virus, a caveolar marker (19). Second, stimulationof caveolar endocytosis occurs when cells are brieflyincubated with either natural or synthetic GSLs such asbovine GM1 ganglioside, bovine LacCer, D-Lactosyl-β1-1′-N-octanoyl-D-erythro-sphingosine (C8-D-erythro-LacCer),or Bodipy-LacCer (9,20). A third important factor forcaveolar endocytosis is that this process is regulatedby microdomain clustering. This notion is supported bythe findings that (i) treatments that promote clusteringof cargo (e.g. SV40 virus, CtxB subunit crosslinking,D-erythro-LacCer) in microdomains can also stimulatecaveolar endocytosis (4,9,21,22). In contrast, treatmentsthat inhibit clustering in microdomains (e.g. methyl-β-cyclodextrin, filipin, C8-L-threo-LacCer) inhibit endocytosisvia caveolae (9,11,21,23).

In the current study, we examine the role of endogenouscell surface gangliosides and show that they also play akey role in regulation of caveolar uptake. Gangliosides

348 www.traffic.dk

Gangliosides, Caveolae and Focal Adhesions

are sialylated GSLs; they are highly diverse becauseof the various combinations of carbohydrate buildingblocks, linkage position, number and position of sialicacid residues, and the precise molecular structure ofthe ceramide lipid anchor (24). In addition to serving asreceptors for certain viruses and toxins, gangliosideshave also been shown to participate in a number ofcomplex cellular processes including cell–cell interactionsand regulation of certain growth factor receptors (e.g.epidermal growth factor and the insulin receptor) (25).Importantly gangliosides are not only components ofPM domains including caveolae, but they can interactwith integrins (26,27) sometimes in association with othertransmembrane proteins (e.g. the tetraspanins), and theycan modulate integrin-based cell attachment (28). Herewe demonstrate for the first time that gangliosides are

required for the maintenance of PM caveolae and forcaveolar endocytosis, and show that they regulate theorganization of PM microdomains, β1-integrin activationand focal adhesion assembly.

Results

Cell surface sialic acid residues support caveolar

endocytosis

To study the effects of loss of cell surface sialic acidson endocytic pathways, HSFs were treated with siali-dase from Arthrobacter ureafaciens which cleaves theterminal N-acetylneuraminic acid residues from the cellsurface glycoproteins and gangliosides (29) (Figure S1).After sialidase treatment, cells were incubated with

Figure 1: Effect of sialic acid deple-

tion on endocytosis. A and B) HSFswere preincubated for 30 min at 37◦Cwithout (Control) or with sialidase,trypsin, or EGCase, washed, and theinternalization of Bodipy-LacCer, Tfn,Dextran, GFP–GPI and interleukin-2β

(IL-2β) receptor were assessed after5 min of endocytosis at 37◦C. A) Flu-orescence micrographs showing theeffects of sialidase, trypsin, or EGCasepretreatments on Bodipy-LacCer andTfn uptake. Dotted lines outline thecells in field. B) Uptake of multiplemarkers after 5 min of endocytosis at37◦C. Note the selective inhibition ofendocytosis of Bodipy-LacCer (but notother markers) following pretreatmentwith sialidase or EGCase. Endocytosiswas quantified by image analysis (n ≥30 cells/marker in three independentexperiments) and expressed as means± SE relative to untreated control sam-ples. See Figure S2A for optimizationof sialidase and EGCase pretreatmenttimes on Bodipy-LacCer endocytosis. Cand D) Endocytosis of Bodipy-LacCer,Tfn, and dextran in a mutant CHO cellline that has decreased levels of cell sur-face sialic acid. C) Fluorescence micro-graphs of Bodipy-LacCer and Tfn in theCHO parental (Pro5) and Lec-2 mutantcells (defective in the transport of CMP-NeuAc into the lumen of the Golgiapparatus) after 5 min of endocytosisat 37◦C. D) Quantitation of uptake byimage analysis (n ≥ 30 cells/marker inthree independent experiments). Scalebars, 10 μm.

Traffic 2010; 11: 348–360 349

Singh et al.

various markers for uptake via clathrin-dependent orclathrin-independent endocytosis (1). Pretreatment withsialidase dramatically inhibited internalization of Bodipy-LacCer (Figure 1A; Figure S2A), a marker for caveolarendocytosis (8,9,13–15,30). Lipid extraction and anal-ysis demonstrated that approximately equal amountsof Bodipy-LacCer became cell-associated at 10◦C inuntreated (control) versus sialidase-treated cells, demon-strating that the inhibition of caveolar endocytosis bysialidase was not due to lower levels of PM labelingwith Bodipy-LacCer prior to endocytosis at 37◦C (FigureS2B). In addition, sialidase pretreatment had no effect oninternalization of markers for the clathrin pathway [trans-ferrin (Tfn); Figure 1A], for fluid phase uptake [dextranand glycosylphosphatidylinositol anchored green fluores-cent protein (GFP-GPI)], or RhoA-dependent uptake of theinterleukin-2 receptor beta (IL-2Rβ) (Figure 1B). In contrast,pretreatment with trypsin had no effect on Bodipy-LacCeror dextran internalization but significantly inhibited endo-cytosis of Tfn, GFP–GPI and IL-2Rβ, presumably due todirect proteolytic cleavage of the cell surface proteins (e.g.Tfn receptor, GFP–GPI) being monitored (Figure 1A,B).

To further test the role of cell surface sialic acid residueson endocytosis, we also examined endocytosis in Lec2cells, a mutant Chinese hamster ovary (CHO) cell lineunable to translocate CMP-sialic acid to the lumen ofthe Golgi apparatus and consequently deficient in sialicacid-containing glycoproteins and glycolipids versus theparental CHO Pro5 cell line (31). As shown in Figure 1C,D,there was a drastic reduction in Bodipy-LacCer endocyto-sis in Lec2 cells versus Pro5 cells, whereas Tfn uptakewas unaffected and dextran internalization was somewhatelevated relative to that in the CHO (Pro5) control cells.The strong inhibition of caveolar endocytosis in Lec2 cellsis consistent with the results of sialidase treatment ofHSFs shown in Figure 1A,B.

Because sialidase pretreatment and the Lec2 mutationaffect both cell surface GSLs and glycoproteins, we nextexamined the effect of endoglycoceramidase (EGCase)on endocytosis as this enzyme only modifies GSLs (i.e.hydrolyzes the linkage between the oligosaccharide andceramide of acidic and neutral GSLs; see Figure S1)but not proteins. Similar to the results using sialidase,

Figure 2: Restoration of Bodipy-LacCer endocytosis in HSFs by sialic acid containing gangliosides. Cells were pretreated withsialidase (A and B) or EGCase (C) as in Figure 1, washed, and pulse-labeled with Bodipy-LacCer to monitor uptake after 5 min ofendocytosis at 37◦C. In some instances, the enzyme-pretreated cells were incubated for 30 min at 10◦C with the indicated non-fluorescent lipid, washed and subsequently pulse-labeled with Bodipy-LacCer as above. A) Fluorescence micrographs showing inhibitionof Bodipy-LacCer uptake by sialidase and restoration of endocytosis by GM1 (but not asialo GM1) ganglioside. Scale bar, 10 μm. (Band C) Quantitative results for restoration of Bodipy-LacCer endocytosis by various lipids after sialidase or EGCase treatments. ‘Notreatment’ refers to no enzymatic pretreatment; ‘No lipid’ shows effect of enzymatic treatment without lipid addition. Values werequantified as in Figure 1 (n ≥ 30 cells/marker in three independent experiments) and are expressed as means ± SE relative to untreatedcontrol samples.

350 Traffic 2010; 11: 348–360

Gangliosides, Caveolae and Focal Adhesions

pretreatment of cells with EGCase selectively inhibitedcaveolar endocytosis of Bodipy-LacCer but had no effecton internalization by other mechanisms (Figure 1A,B), sug-gesting that cell surface sialic acid-containing lipids (i.e.gangliosides) rather than proteins might be responsiblefor the observed effects on caveolar endocytosis. Thisinhibition was not due to the formation of ceramideby EGCase, because incubation of untreated cells withC8-ceramide did not significantly inhibit uptake of Bodipy-LacCer (Figure S2C,D). To examine the effect of sialidaseand EGCase treatments on endocytosis of Bodipy-LacCerfurther, we attempted to reverse their effects by incu-bating the enzymatically treated cells with a panel ofGSLs prior to monitoring LacCer endocytosis. Additionof individual gangliosides (i.e. sialic-acid containing GSLssuch as GM1, GM2, GD1a) as well as ganglioside mix-tures restored Bodipy-LacCer endocytosis in sialidase- andEGCase-treated cells, whereas neutral GSLs (e.g. asialoGM1, asialo GM2, C8-D-erythro-LacCer) or neutral GSLmixtures were not able to restore Bodipy-LacCer uptake(Figure 2). Importantly, no restoration of caveolar endo-cytosis was seen when sulfatide (a negatively chargedSL; see Figure S1) was used (Figure 2B), suggesting thatrestoration did not simply result from the presence ofnegatively charged lipids at the PM. Together the resultsin Figures 1 and 2 demonstrate that sialidase and EGCaseexert their effects on endocytosis by altering cell surfacegangliosides and that these lipids are critical in supportingcaveolar endocytosis.

Effect of sialidase and EGCase treatments on

caveolae and Cav1

We examined the effect of sialidase and EGCasetreatments on PM caveolae by transmission electronmicroscopy. For these studies, samples were preparedusing Ruthenium red to identify caveolae connected to thecell surface as described (32). The number of caveolae,defined as uncoated surface invaginations between 50and 80 nm in diameter, was counted in a blindedmanner. As shown in Figure 3, control HSFs showednumerous caveolae on both the dorsal (heavily stained)and ventral (lightly stained) surfaces of the PM whereastreatment with sialidase or EGCase for 30 min at 37◦C(identical conditions to those used in Figures 1 and 2)reduced the number of PM caveolae by approximately80%.

We next used immunofluorescence to examine theeffects of sialidase and EGCase pretreatments on theintracellular distribution of Cav1 using an antibody andfixation-permeabilization protocol to stain endogenousCav1 pools at both the Golgi apparatus and cellsurface (33). When control cells were examined by totalinternal reflection fluorescence (TIRF) microscopy, strongstaining at the PM was seen while in sialidase- andEGCase-treated cells, Cav1 staining of the cell surfacewas significantly reduced (Figure 4A). A similar result wasobtained using another antibody and staining protocol

Figure 3: Effect of sialidase and EGCase treatments on

caveolae. A) HSFs were untreated (Control) or treated for 30 minat 37◦C with sialidase or EGCase, fixed, stained with rutheniumred to identify PM invaginations, and processed for transmissionelectron microscopy. Scale bar, 500 nm. B) For quantitation,50–80-nm diameter smooth, surface-connected vesicles within0.5 μm of the cell surface were counted along the entire cellperimeter and are expressed (means ± SE) as number per 100 μmof length.

Traffic 2010; 11: 348–360 351

Singh et al.

Figure 4: Cav1 is lost from

the PM following sialidase

or EGCase treatments but is

restored by incubation with

exogenous GM1 ganglioside at

low temperature. A) HSFs wereuntreated or treated with the indi-cated enzyme for 30 min at 37◦C,fixed, immunostained for Cav1,and observed by epifluorescenceor TIRF microscopy. Note theloss of Cav1 from the cell sur-face (TIRF images) in the treatedcells. B) Real time imaging ofCav1–GFP demonstrates redistri-bution of Cav1 from the cell sur-face to intracellular membranesduring incubation with sialidase at37◦C. Indicated time points aretaken from QuickTime movies.Control experiments using bufferalone showed no redistribution ofCav1–GFP over the same timeperiod. C) Total Cav1 levels bywestern blotting were not alteredby sialidase or EGCase treatments.D and E) Endogenous Cav1 redis-tributes back to the PM of EGCase-treated cells following a 30 minincubation with exogenous GM1

ganglioside at 10◦C. Quantitationof TIRF signal for Cav1 in panel(E) was by image analysis (n ≥ 30cells/condition) and is expressed(mean ± SE) as a percent of con-trols. Scale bars, 10 μm.

specific for the PM pool of Cav1 (33) (data not shown).Furthermore, real time (TIRF) imaging of HSFs expressingCav1–GFP showed a rapid decrease in Cav1 at the PMduring the first few minutes of sialidase treatment at37◦C, whereas in untreated control cells, no changes wereobserved over the same time period (Figure 4B). Westernblotting of cells further demonstrated that the totalendogenous Cav1 pool was not diminished by the enzymetreatments (Figure 4C), and epifluorescence images ofthe cells demonstrated that Cav1 had redistributed intocytoplasmic pools. Furthermore, incubation of EGCase-treated cells with exogenous GM1 ganglioside for 30 minat 10◦C (identical conditions to those in Figure 2 thatrestored caveolar endocytosis) induced endogenous Cav1to redistribute back to the PM (Figure 4D,E).

In agreement with published studies, we also found thatPTRF-Cavin (5,6), a caveolar protein, colocalized with Cav1at the PM in control cells (Figure 5). However, in sialidaseand EGCase treated cells, Cav1 was redistributed tothe Golgi apparatus while PTRF-Cavin showed a non-overlapping diffuse and punctate distribution in the cells(Figure 5). Taken together, the results in Figures 3–5demonstrate that sialidase and EGCase treatmentsreduced PM caveolae and caused a redistribution of keyproteins normally associated with caveolae, providing astructural basis for the observed inhibition of caveolarendocytosis. Importantly, exogenous sialic acid-containinggangliosides restored the distribution of Cav1 at the PMin the enzyme-treated cells (Figure 4D,E) providing a basisfor restoration of caveolar endocytosis in these cells.

352 Traffic 2010; 11: 348–360

Gangliosides, Caveolae and Focal Adhesions

Figure 5: Effect of sialidase or

EGCase treatment on the distri-

bution of Cav1 and PTRF-Cavin.

HSFs were treated with or withoutthe indicated enzyme for 30 minat 37◦C followed by immunostain-ing of Cav1 (red) and PTRF-Cavin(green). Note the typical surface-like staining of Cav1 and extensivecolocalization with PTRF-Cavin incontrol cells, while in sialidase-or EGCase-treated cells these pro-teins were non-overlapping and indifferent structures. Arrows indi-cate Golgi region. Scale bars,10 μm.

Disruption of PM domains by sialidase and EGCase

treatment

We have shown that clustering of lipids and proteins intoGSL- and cholesterol-enriched micron-size domains (or‘rafts’) at the PM can be monitored using Bodipy-LacCer(reviewed in 34), and that such domains are requiredfor caveolar endocytosis in HSFs (21). Thus, we nextsought to learn whether these PM microdomains werealtered by sialidase and EGCase treatments. To do so,after sialidase or EGCase pretreatments, cells were incu-bated with Bodipy-LacCer for 30 min at 10◦C, and live cellimages were acquired at 10◦C to inhibit endocytosis. Todistinguish PM domains enriched in LacCer from otherregions of the PM containing lower concentrations of thelipid analog, we simultaneously monitored both monomer(green; low concentration) and excimer (red; high con-centration) fluorescence [Figure 6 and Refs (34,35)]. Incontrol cells, Bodipy-LacCer was enriched in numerousmicron-size ‘clusters’ which exhibited yellow/orange flu-orescence, while neighboring regions of the cell surfacethat contained less of the lipid analog fluoresced green. Incontrast, pretreatment of cells with EGCase or sialidaseresulted in loss of the yellow/orange clusters of fluores-cence at the PM indicating a reduction or loss of micron-size PM clusters (Figure 6; Figure S3A). As noted above,lipid extraction and analysis further demonstrated that theamount of fluorescent LacCer that became cell-associatedat 10◦C was not affected by pretreatment of HSFs withsialidase or EGCase (Figure S2B). Importantly, when theenzymatically treated cells were subsequently incubatedwith GM1 ganglioside at 10◦C followed by incubation withBodipy-LacCer, the micron-size clusters of yellow/orangefluorescence at the PM were again visible (Figure 6). How-ever, this restoration of PM domains was not observed

when cells were incubated with asialo GM1 ganglioside(Figure 6A,B). These experiments demonstrate that thepresence of sialic acid residues on lipids is required forPM domains as monitored by clustering of Bodipy-LacCer.We also used PEG–cholesterol as another marker for PMdomains (36) and found that this probe was (i) enrichedin micron-size clusters at the PM of control cells similarin appearance to those seen using Bodipy-LacCer; and (ii)these clusters were lost when cells were pretreated withEGCase (Figure S3B,C).

Effect of EGCase and Sialidase treatments on integrin

activation and focal adhesions

In a previous study, we showed that a non-naturalstereoisomer of LacCer inhibits caveolar endocytosis,blocks the clustering of lipids and proteins into GSL- andcholesterol-enriched PM domains, and blunts β1-integrinactivation (21). Because these first two effects were seenin HSFs treated with sialidase or EGCase (Figures 1 and6), we next examined the effects of EGCase treatmenton β1-integrin activation and on the distribution of severalcomponents of the focal adhesion complex (Figures 7and 8). To examine integrin activation, we used HUTS-4, a monoclonal antibody that binds to β1-integrins intheir activated conformation. Untreated cells grown inserum-containing medium showed a basal level of integrinactivation that was significantly reduced in EGCase-treated cells (Figure 7A,B). However, EGCase treatmentdid not alter the PM levels of β1-integrin detected byimmunofluorescence (Figure 7C,D) or the total levelsof β1-integrin detected by western blotting (Figure 7E).Importantly, this inhibition of integrin activation waspartially restored when EGCase treated cells were

Traffic 2010; 11: 348–360 353

Singh et al.

Figure 6: Enzymatic cleavage of cell surface sialic acid

residues abolished PM domains enriched in Bodipy-LacCer.

HSFs were untreated (Control), or pretreated with EGCase (A andB) or sialidase (C) for 30 min at 37◦C, followed by incubation with2.5 μM Bodipy-LacCer for 30 min at 10◦C. Cells were then washedand fluorescence images were acquired simultaneously at greenand red wavelengths and merged. Imaging was carried out atlow temperature to inhibit endocytosis. A) Note the presenceof yellow/orange micron-size clusters in control cells (e.g. atarrows) whereas no such domains were observed in EGCase-treated cells. PM domain formation was restored when theenzyme-treated cells were incubated for 30 min at 10◦C withexogenous GM1 ganglioside prior to incubation with Bodipy-LacCer. No restoration of PM domains was seen using asialoGM1 in place of GM1. Scale bars, 10 μm. See Figure S3 for animage of BODIPY-LacCer-labeled sialidase-treated cells. In (B andC) the number of ‘patches’ ≥ 1 × 1 μm were quantified by imageanalysis and are expressed per 100 μm2 area (n ≥ 6 cells percondition in two or more independent experiments).

incubated with exogenous GM1 ganglioside (Figure 7A,B).Treatment with EGCase caused a redistribution of focaladhesion kinase (FAK), pFAK, Type Iγ phosphatidylinositolphosphate kinase (PIPKIγ), paxillin and talin away fromfocal adhesions (Figure 8A,B), while no change in thetotal cellular levels of each of these proteins wasseen by western blotting (Figure 8C). Together thedata in Figures 7 and 8 demonstrate that EGCasetreatment inhibited β1-integrin activation, and disruptedthe localization of several key proteins found in focaladhesion complexes.

Knockdown of β1-integrin

In a previous study using HSFs, we showed that depletionof β1-integrin inhibited clustering and activation of thisintegrin and blocked caveolar endocytosis (21). Becausepretreatment of HSFs with EGCase also blunts activationof β1-integrin and inhibits caveolar endocytosis, we nextexamined the effect of β1-integrin knockdown on PMdomains and on the distribution of Cav1. Using the siRNAapproach (21), we depleted β1-integrin >95% (Figure S4).No effects of β1-integrin knockdown on cell attachmentwere observed, presumably because of the presence ofother integrins and other adhesion proteins in the HSFs.We then incubated these cells with Bodipy-LacCer orAF647 PEG–Chol to observe the effect of the knockdownon PM domains. Micron-size clusters of Bodipy-LacCer orPEG–Chol were readily apparent in untreated control cellsand in cells transfected with a negative siRNA, but werealmost completely eliminated in β1-integrin-depleted cellsusing either domain marker (Figure 9A–C). β1-integrindepletion also caused a redistribution of Cav1 from thePM to intracellular membranes (Figure 9D,E). The lossof micron-size PM domains and redistribution of Cav1away from the PM in β1-integrin-depleted cells (Figure 9)were similar to the effects seen when HSFs were treatedwith EGCase (Figures 4 and 6). These results suggestthat β1-integrins are required for the formation of PMdomains and Cav1 association with the PM. However,incubation with exogenous GM1 ganglioside at 10◦C didnot induce the redistribution of Cav1 back to the PM(Figure 9D,E), suggesting that both gangliosides and β1-integrin are required for this process to occur.

Discussion

In the present study, we found that enzymatic deple-tion of cell surface sialic acid residues using sialidase orEGCase inhibited caveolar endocytosis in HSFs, but didnot affect other mechanisms of internalization. Caveolarendocytosis was also inhibited in a mutant CHO cell linedeficient in the synthesis of sialic acid-containing glycopro-teins and glycolipids, suggesting that the effects reportedhere are not limited to HSFs. We were able to attributethis inhibition to depletion of gangliosides, rather than aloss of sialic acid residues from glycoproteins for severalreasons. First, when cells were incubated with EGCasewhich modifies cell surface GSLs but not proteins (37,38),

354 Traffic 2010; 11: 348–360

Gangliosides, Caveolae and Focal Adhesions

Figure 7: Effect of EGCase treatment on β1-integrin levels and activation. HSFs were grown in serum-containing medium, washedand were either untreated (Control) or pretreated with EGCase for 30 min at 37◦C. A and B) Control or EGCase treated cells werewashed, warmed for 30 sec at 37◦C, fixed and immunostained with HUTS-4 mAb to detect activated β1-integrin. In one experimentthe EGCase treated cells were subsequently washed and incubated for 30 min at 10◦C with GM1 ganglioside prior to HUTS-4 fixationand staining. In (B), images from the experiment in panel (A) were quantified as in panel by image analysis (n ≥ 50 cells) and areexpressed as a percent of untreated control samples (mean ± SE). C and D) Live cells were stained for 30 min at 10◦C using aβ1-integrin FITC-labeled antibody (monitors both activated and non-activated β1-integrin), washed and observed under the fluorescencemicroscope. No significant difference in fluorescence intensity between Control and EGCase treated cells (n ≥ 50 cells/condition) wasobserved. E) Cells treated as in (C and D) were lysed and blotted for β1-integrin (10 μg protein per lane). Scale bars, 10 μm.

virtually identical results were obtained to those foundusing sialidase [e.g. on endocytosis (Figure 1), or on theloss of PM caveolae and Cav1 (Figures 3 and 4)]. Sec-ond, the effects of sialidase treatments on endocytosiscould be reversed by exogenous GM1 and other gan-gliosides, but not by asialo–gangliosides or neutral GSLs(Figure 2). It is important to note that cell surface gan-gliosides have been shown to act as receptors for theuptake of certain pathogens [e.g. SV40 virus and Choleratoxin and Mycoplasma pneumoniae; (4,39–41)]. Subse-quent entry of many of these pathogens usually occurs bya clathrin-independent mechanism such as caveolar endo-cytosis (40,42,43) and perturbation of the PM domains(e.g. by cholesterol depletion) can inhibit internalization ofthe pathogen (44). However, to our knowledge, our studyis the first to show that PM gangliosides are required forcaveolar endocytosis as well as the maintenance of PMCav1 and caveolae (Figures 1, 3 and 4).

Treatment of HSFs with sialidase or EGCase inhibitedcaveolar endocytosis (Figure 1), but also had other effectsthat may be related to blocking this endocytic mechanism.These include (i) inhibition of PM domains, (ii) loss of Cav1from the PM and (iii) inhibition of β1-integrin activation

(Figures 4, 6 and 7). Each of these effects could bereversed by incubation with an exogenous ganglioside. Amajor open question regarding the underlying mechanismfor ganglioside regulation of caveolar endocytosis iswhether gangliosides affect each of these processesindependently, or whether there is a temporal sequenceof events in which ganglioside depletion inhibits oneprocess and disruption of that process in turn blocksother processes leading to the eventual inhibition ofcaveolar endocytosis. For example, the primary effectof ganglioside depletion could be to block PM domainformation with the absence of PM domains preventingactivation of β1-integrin and the formation of caveolae.Another possibility is that the loss of gangliosidestransiently stimulates the endocytosis of caveolae andrelated microdomains at the PM, similar to the effects ofcell detachment on internalization of caveolae reportedby del Pozo et al. (45) These issues may be furtherexplored in future studies by treating cells with sialidaseor EGCase at various (low) concentrations or for varioustimes to achieve different levels of ganglioside depletion.For a given level of depletion, temporal changes in PMdomains, activation of β1-integrin and loss of endogenousCav1 from the PM could then be examined. Results from

Traffic 2010; 11: 348–360 355

Singh et al.

Figure 8: Effect of EGCase treat-

ment of HSFs on focal adhe-

sion components. A) Cellswere grown in serum-containingmedium, washed and subse-quently were untreated (Control)or pretreated with EGCase for30 min at 37◦C. The basal levelof FAK and pFAK (Y397) wasthen assessed by antibody stain-ing using both epifluorescence andTIRF microscopy. B) Parallel cellsamples, grown under the sameconditions, were immunostainedfor talin or paxilin. In the caseof PIPK1γ cells were transfectedovernight with HA-tagged PIPKIγprior to EGCase treatment, fixa-tion and immunostaining. PIPKIγand talin images were by confocalmicroscopy; paxilin images wereby TIRF microscopy. C) In paralleldishes, cells were untreated (Con-trol) or pretreated with EGCase for30 min at 37◦C, lysed, and celllysates (10 μg protein per lane)were immunoblotted for variousproteins as indicated. Scale bars,10 μm.

these studies may indicate whether these processes aresequential (e.g. does loss of PM domains precede loss ofCav1 from the PM) and possibly linked, or whether all ofthese events occur simultaneously suggesting they maybe independent of one another.

In summary, this study demonstrates that PM ganglio-sides can selectively regulate caveolar endocytosis andrelated processes in HSFs. The regulation of caveolaeand microdomains by gangliosides could be physiolog-ically important as there are many reports of changesin gangliosides in normal or pathological states. Forexample, endogenous sialidases (e.g. Neu3) which playkey roles in cancer and differentiation (46) could functionby regulating microdomain based signaling. In addition,certain tumor cells express aberrant levels and speciesof gangliosides (47,48) and shed these gangliosides intothe tumor microenvironment thus affecting surroundingcells (49,50). Such changes in gangliosides could affecttumor invasiveness, focal adhesion formation and angio-genesis by modulating the formation of caveolae orintegrin signaling.

Materials and Methods

Cell cultureNormal HSFs (GM-5659; Coriell Institute for Medical research) were grownas described (21). CHO mutant (Lec-2) and parental (Pro-5) cell lines werepurchased from ATCC, and maintained per instructions supplied by theATCC. All experiments were performed using monolayer cultures grownto approximately 40–60% confluency on acid-etched glass coverslips.

Lipids, fluorescent probes and miscellaneous

reagentsBodipy-D-e-LacCer was synthesized, purified and subsequently complexedto defatted bovine serum albumin (DF-BSA) for incubation with cells (30).Fluorescent Alexa Fluor (AF) 594 labeled-transferrin (Tfn), AF594-dextran(10kD), Ab labeling kits, and fluorescent secondary Abs were fromInvitrogen. Anti GFP (IgG1) antibodies were from Sigma. Anti GFP Fabfragments were generated from this IgG1 using the Immunopure IgG1Fab preparation kit from Pierce Biotechnology and were labeled withAF647 succinimidyl ester using a protein labeling kit from Invitrogen.PIPLCγ (Phosphatidylinositol phospholipase Cγ) was purchased fromSigma Chemical Co. Antibodies were purchased from BD Biosciences(β1-integrin, Paxillin, FAK, and pFAK), Millipore (HUTS-4 β1-integrin), andSigma Chemical Co. (Talin). Anti PIPK1γ was obtained as described (51).Sialidase was from Sigma Chemical Co. and Glyko Biochemicals; EGCase

356 Traffic 2010; 11: 348–360

Gangliosides, Caveolae and Focal Adhesions

Figure 9: Effect of β1-integrin

knockdown on PM domains and

the distribution of Caveolin-1.

HSFs were transfected for 72 hwith negative (Control) or β1-integrin siRNA (Figure S4). (A–C)Cells were then incubated withBodipy-LacCer or AF647-PEG-Cholto visualize PM clusters (e.g. atarrows of Control samples). In (Band C) the number of micron-size clusters of Bodipy-LacCer orPEG–Chol were quantified as inFigure 6. D and E) Cells as abovewere fixed and immunostained forCav1. Samples were then visual-ized by TIRF or epifluorescencemicroscopy. In (E) TIRF imagesfrom the experiment in panel (D)were quantified by image analysis(n ≥ 50 cells) and are expressedas a percent of untreated controlsamples (mean ± SE). Scale bars,10 μm.

from Takara Biochemicals and Sigma Chemical Co. All other reagents werefrom Sigma-Aldrich.

Constructs and transfection experimentsDNA constructs encoding GPI-GFP and interleukin-2 receptor (IL-2R) weregifts from J. Lippincott-Schwartz (NIH) and A. Dautry-Varsat (InstitutPasteur), respectively. HA-tagged PIPKIγ was described previously (51).Transfection of DNA constructs was performed using a Nucleofector IIapparatus (Amaxa Biosystems).

β1-integrin knockdownsiRNAs for β1-integrin (ON-TARGETplus SMARTpool L-004506-00) andnegative siRNA (ON-TARGETplus Non-targeting siRNA D-001810-01-05)were purchased from Thermo Scientific. Knockdown in HSFs wasperformed using 360 pmol/mL siRNA with DharmaFECT 2 transfectionreagent (Thermo Scientific) as per supplier’s instructions. Knockdownwas confirmed by western blotting and immunofluorescence 72 h aftertransfection as shown in Figure S4. All further experiments were alsoperformed 72 h after transfection.

Enzymatic treatments of intact cellsHSFs were pretreated with sialidase (0.1 unit/mL medium) or EGCase (20milliunits/mL medium) for 30 min at 37◦C at pH 5.5, washed and returnedto pH 7.2 medium. Control cells were treated identically at the same pHwithout enzyme. Cells were then washed and pH adjusted to 7.2 forfurther binding and endocytosis assays. In some experiments, cells werepretreated with 0.01% trypsin (Trypsin type II tablets; Sigma Chem. Co) for30 min at 37◦C, followed by 5% BSA for inactivation of the enzyme. Cellswere then washed and further used for various endocytosis or signalingassays as described elsewhere in Materials and Methods. No apparenteffect on cell viability was seen for any of these treatments.

Endocytosis assaysFor endocytosis of Bodipy-LacCer or AF labeled Tfn, HSFs uponpretreatments (with either enzymes or with trypsin) were incubatedwith the fluorescent marker for 30 min at 10◦C, washed and furtherincubated for 3 min at 37◦C (21). For AF labeled dextran, cells wereincubated for 5 min at 37◦C without preincubation after the pretreatment.Endocytosis of IL-2R was performed as described previously (14). For GPI-anchored protein uptake, HSFs were first transfected with GPI-GFP for48 h and then were pretreated as above. Cells were then incubated withAF647-labeled anti-GFP-Fab for 30 min at 10◦C, washed and incubated for

Traffic 2010; 11: 348–360 357

Singh et al.

5 min at 37◦C. Cells were then treated with PIPLCγ (Phosphatidylinositolphospholipase Cγ) (Sigma Chemical Co) to remove cell surface GPI-GFP.All other samples were either acid stripped or back exchanged to removecell surface fluorescence before fluorescence microscopy. Endocytosis ofBodipy-LacCer, AF labeled Tfn and dextran was carried out identically inmutant Lec-2 and their parental cell line Pro-5.

Electron microscopyFor EM studies, HSFs were grown on Aclar sheets placed in 35-mmculture dishes and were either untreated (control) or treated with sialidaseor EGCase for 30 min at 37◦C (see above). Samples were then fixed inthe presence of 1 mM ruthenium red and embedded as described (20).Transverse ultrathin sections were cut and viewed under an FEI TecnaiT12 transmission electron microscope operating at 80 kV. Overlappingimages of the entire cell perimeter in a given field were taken at × 15000 magnifications. For quantitation, the number of ruthenium red positive,50–80 nm-diameter vesicles within 0.5 μm of the cell surface was countedfor the entire cell perimeter. Values are expressed per 100 μm of perimeterlength.

Microdomain studiesHSFs were washed with ice cold HMEM and were pretreated withor without sialidase or EGCase for 30 min at 37◦C. Cells were thenwashed and incubated with 2.5 μM Bodipy-LacCer for 30 min at 10◦C tolabel the PM. Samples were then washed, and images were acquiredsimultaneously at green and red wavelengths using a Dual View module(Optical Insights). For glycolipid supplementation, cells were incubated withGM1(as described below) for 30 min at 10◦C after enzyme pretreatment.Cells were then washed and further incubated with 2.5 μM Bodipy-LacCerfor 30 min at 10◦C, washed and imaged as above.

Glycolipid supplementation assayDifferent lipids (20 μg/mL), as indicated, were dissolved in EtOH:DMSO ::1:1 and were added in HMEM +G and incubated after differentpretreatments as indicated for 30 min at 10◦C. Cells were then washedthree times and processed for further endocytosis, microdomains orrestoration of Cav1 at the PM.

Fluorescence microscopy and analysisFluorescence microscopy was performed using a fluorescence microscope(IX70; Olympus) equipped with × 60 1.4NA or × 100 1.35 NA oil immersionobjectives. Images were acquired using a Quant EM: 512SC (Photometrics)CCD camera. For quantitation, all photomicrographs in a given experimentwere exposed and processed identically for a given fluorophore and wereanalyzed using the MetaMorph image processing program (version 7.3.2;Universal imaging Corp.). Quantitative results are expressed as mean ±SDs. Images were prepared for individual figures using Photoshop CS(Adobe). Total internal reflection fluorescence (TIRF) microscopy wascarried out using an Olympus attachment for the I×70 microscope.No deconvolution, 3D reconstructions, surface or volume rendering, orgamma adjustments were performed. Some micrographs in Figure 8 wereobtained using a Zeiss LSM510 confocal microscope equipped with × 1001.3NA lens. Details regarding fluorophores and temperature are given inthe text or figure legends.

Miscellaneous proceduresSDS-PAGE and immunoblotting of cell lysates and immunofluorescenceof formaldehyde- or methanol-fixed cells were performed as describedpreviously (21,33). Lipid extraction and thin layer chromatography analysis(TLC) were performed as described (52). SLs were separated by TLC andidentified by comparison to known standards using CHCl3/CH3OH/15 mM

CaCl2 (65:35:8; v/v/v) as the developing solvent. Primulin or resorcinolwas used as a detection reagent, and lipids were quantified by scanningdensitometry (53).

Acknowledgments

This work was supported by the Mayo Foundation and grants from theNIH Grants GM-22942 and GM-60934 to REP.

Supporting Information

Additional Supporting Information may be found in the online version ofthis article:

Figure S1: Sialidase or EGCase cleavage of GSLs. A) Carbohydratemoieties of several different sphingolipids showing sites of cleavage bythe indicated enzymes. B) TLC analysis of HSF lipids before and aftertreatment with Sialidase for 30 min at 37◦C (see Materials and Methods).Sialylated GSLs were selectively stained using resorcinol. GM3 and twoother gangliosides (X and Y) are indicated. Image analysis indicated thatGM3 ganglioside was reduced approximately 45% by sialidase. Primulinstaining of the same samples demonstrated the formation of LacCerfollowing sialidase treatment, as expected (see panel A).

Figure S2: Effect of sialidase, EGCase, or C8-ceramide on Bodipy-

LacCer incorporation and endocytosis in HSFs. A) Cells were pretreatedwith the indicated enzyme for 0–30 min at 37◦C, washed, pulse-labeledwith Bodipy-LacCer (as in Figure 1), and uptake (after 5 min at 37◦C)was quantified by image analysis. Values are mean ± SD (n > 40cells in 2 independent experiments). B) Cells were untreated (Control)or pretreated with sialidase or EGCase for 30 min at 37◦C, washed,incubated with Bodipy-LacCer for 30 min at 10◦C, washed, and the lipidsextracted. The amount of cell-associated Bodipy-LacCer was quantified ina fluorometer and normalized to untreated control cells. C and D) HSFswere incubated without (Control) or with 20 μM C8-ceramide for 30 minat 10◦C, washed and then pulse-labeled with Bodipy-LacCer as in (A). D)Images corresponding to panel (C) were quantified as in Figure 1. Scalebar, 10 μm.

Figure S3: Effect of enzymatic treatments on PM domains. HSFs wereuntreated (Control), or pretreated with sialidase or EGCase as indicated for30 min at 37◦C. The samples were subsequently incubated for 30 min at10◦C with (A) Bodipy-LacCer as in Figure 6A, or B 1 μM AF647 PEG–Chol,washed, and fluorescence images of living cells acquired at 10◦C. C)Images corresponding to panel (B) were quantified as in Figure 6. Scalebars, 10 μm.

Figure S4: Knockdown of β1-integrin in HSFs. HSFs were transfectedwith negative siRNA (Control) or β1-integrin siRNA for 72 h. A) Cell lysates(10 μg protein per lane) were prepared and samples were blotted for β1-integrin, Cav1, and β-actin. B) Cells were stained by immunofluorescencefor β1-integrin. Approximately 90% of the cells were depleted of β1-integrinas judged by immunofluorescence. Dashed lines outline dim cells in field.Scale bar, 10.

Please note: Wiley-Blackwell are not responsible for the content orfunctionality of any supporting materials supplied by the authors.Any queries (other than missing material) should be directed to thecorresponding author for the article.

References

1. Mayor S, Pagano RE. Pathways of clathrin-independent endocytosis.Nat Rev Mol Cell Biol 2007;8:603–612.

2. Doherty GJ, McMahon HT. Mechanisms of endocytosis. Annu RevBiochem 2009;78:857–902.

3. Hansen CG, Nichols BJ. Molecular mechanisms of clathrin-indepen-dent endocytosis. J Cell Sci 2009;122:1713–1721.

4. Pelkmans L, Helenius A. Endocytosis via caveolae. Traffic 2002;3:311–320.

358 Traffic 2010; 11: 348–360

Gangliosides, Caveolae and Focal Adhesions

5. Hill MM, Bastiani M, Luetterforst R, Kirkham M, Kirkham A, Nixon SJ,Walser P, Abankwa D, Oorschot VM, Martin S, Hancock JF, Par-ton RG. PTRF-Cavin, a conserved cytoplasmic protein required forcaveola formation and function. Cell 2008;132:113–124.

6. Liu L, Pilch PF. A critical role of cavin (polymerase I and transcriptrelease factor) in caveolae formation and organization. J Biol Chem2008;283:4314–4322.

7. Schnitzer JE, Oh P, Pinney E, Allard J. Filipin-sensitive caveolae-mediated transport in endothelium: reduced transcytosis, scavengerendocytosis, and capillary permeability of select macromolecules. JCell Biol 1994;127:1217–1232.

8. Singh RD, Puri V, Valiyaveettil JT, Marks DL, Bittman R, Pagano RE.Selective caveolin-1-dependent endocytosis of glycosphingolipids.Mol Biol Cell 2003;14:3254–3265.

9. Sharma DK, Brown JC, Choudhury A, Peterson TE, Holicky E, MarksDL, Simari R, Parton RG, Pagano RE. Selective stimulation of caveolarendocytosis by glycosphingolipids and cholesterol. Mol Biol Cell2004;15:3114–3122.

10. Pelkmans L, Kartenbeck J, Helenius A. Caveolar endocytosis ofsimian virus 40 reveals a new two-step vesicular-transport pathwayto the ER. Nature Cell Biol 2001;3:473–483.

11. Orlandi PA, Fishman PH. Filipin-dependent inhibition of cholera toxin:evidence for toxin internalization and activation through caveolae-likedomains. J Cell Biol 1998;141:905–915.

12. Torgersen ML, Skretting G, van Deurs B, Sandvig K. Internalizationof cholera toxin by different endocytic mechanisms. J Cell Sci2001;114:3737–3742.

13. Puri V, Watanabe R, Singh RD, Dominguez M, Brown JC, Wheat-ley CL, Marks DL, Pagano RE. Clathrin-dependent and -independentinternalization of plasma membrane sphingolipids initiates two Golgitargeting pathways. J Cell Biol 2001;154:535–547.

14. Cheng ZJ, Singh RD, Sharma DK, Holicky EL, Hanada K, Marks DL,Pagano RE. Distinct mechanisms of clathrin-independent endo-cytosis have unique sphingolipid requirements. Mol Biol Cell2006;17:3197–3210.

15. Kumari S, Mayor S. ARF1 is directly involved in dynamin-independentendocytosis. Nat Cell Biol 2008;10:30–41.

16. Parton RG, Joggerst B, Simons K. Regulated internalization ofcaveolae. J Cell Biol 1994;127:1199–1215.

17. Tagawa A, Mezzacasa A, Hayer A, Longatti A, Pelkmans L, Hele-nius A. Assembly and trafficking of caveolar domains in the cell:caveolae as stable, cargo-triggered, vesicular transporters. J Cell Biol2005;170:769–779.

18. Minshall RD, Sessa WC, Stan RV, Anderson RG, Malik AB. Caveolinregulation of endothelial function. Am J Physiol Lung Cell Mol Physiol2003;285:L1179–1183.

19. Pelkmans L, Fava E, Grabner H, Hannus M, Habermann B, Krausz E,Zerial M. Genome-wide analysis of human kinases in clathrin- andcaveolae/raft-mediated endocytosis. Nature 2005;436:78–86.

20. Kirkham M, Fujita A, Chadda R, Nixon SJ, Kurzchalia TV, Sharma DK,Pagano RE, Hancock JF, Mayor S, Parton RG. Ultrastructural identifi-cation of uncoated caveolin-independent early endocytic vehicles. JCell Biol 2005;168:465–476.

21. Singh RD, Holicky EL, Cheng Z, Kim SY, Wheatley CL, Marks DL,Bittman R, Pagano RE. Inhibition of caveolar uptake, SV40 infection,and ®1-integrin signaling by a non-natural glycosphingolipid stereoiso-mer. J Cell Biol 2007;176:895–901.

22. Escriche M, Burgueno J, Ciruela F, Canela EI, Mallol J, Enrich C,Lluis C, Franco R. Ligand-induced caveolae-mediated internalizationof A1 adenosine receptors: morphological evidence of endosomalsorting and receptor recycling. Exp Cell Res 2003;285:72–90.

23. Upla P, Marjomaki V, Kankaanpaa P, Ivaska J, Hyypia T, van derGoot FG, Heino J. Clustering induces a lateral redistribution of α2β1integrin from membrane rafts to caveolae and subsequent PKC-dependent internalization. Mol Biol Cell 2004;15:625–636.

24. Kolter T, Proia RL, Sandhoff K. Combinatorial ganglioside biosynthe-sis. J Biol Chem 2002;277:25859–25862.

25. Hakomori SI. Inaugural article: the glycosynapse. Proc Natl Acad SciU S A 2002;99:225–232.

26. Wang X, Sun P, Al-Qamari A, Tai T, Kawashima I, Paller AS. Carbohy-drate-carbohydrate binding of ganglioside to integrin alpha(5) modu-lates alpha(5)beta(1) function. J Biol Chem 2001;276:8436–8444.

27. Zheng M, Fang H, Tsuruoka T, Tsuji T, Sasaki T, Hakomori S. Regu-latory role of GM3 ganglioside in alpha 5 beta 1 integrin receptorfor fibronectin-mediated adhesion of FUA169 cells. J Biol Chem1993;268:2217–2222.

28. Todeschini AR, Hakomori S. Functional role of glycosphingolipidsand gangliosides in control of cell adhesion, motility, andgrowth, through glycosynaptic microdomains. Biochim Biophys Acta2008;1780:421–433.

29. Iwamori M, Ohta Y, Uchida Y, Tsukada Y. Arthrobacter ureafacienssialidase isoenzymes, L, M1 and M2, cleave fucosyl GM1. GlycoconjJ 1997;14:67–73.

30. Singh RD, Liu Y, Wheatley CL, Holicky EL, Makino A, Marks DL,Kobayashi T, Subramaniam G, Bittman R, Pagano RE. Caveolar endo-cytosis and microdomain association of a glycosphingolipid analogis dependent on its sphingosine stereochemistry. J Biol Chem2006;281:30660–30668.

31. Stanley P, Ioffe E. Glycosyltransferase mutants: key to new insightsin glycobiology. FASEB J 1995;9:1436–1444.

32. Parton RG, Molero JC, Floetenmeyer M, Green KM, James DE.Characterization of a distinct plasma membrane macrodomain indifferentiated adipocytes. J Biol Chem 2002;277:46769–46778.

33. Pol A, Martin S, Fernandez MA, Ingelmo-Torres M, Ferguson C,Enrich C, Parton RG. Cholesterol and fatty acids regulate dynamiccaveolin trafficking through the Golgi complex and between the cellsurface and lipid bodies. Mol Biol Cell 2005;16:2091–2105.

34. Marks DL, Bittman R, Pagano RE. Use of Bodipy-labeled sphingolipidand cholesterol analogs to examine membrane microdomains in cells.Histochem Cell Biol 2008;130:819–832.

35. Pagano RE, Martin OC, Kang HC, Haugland RP. A novel fluorescentceramide analogue for studying membrane traffic in animalcells: accumulation at the Golgi apparatus results in alteredspectral properties of the sphingolipid precursor. J Cell Biol1991;113:1267–1279.

36. Takahashi M, Murate M, Fukuda M, Sato SB, Ohta A, Kobayashi T.Cholesterol controls lipid endocytosis through Rab11. Mol Biol Cell2007;18:2667–2677.

37. Ito M, Komori H. Homeostasis of cell-surface glycosphingolipid con-tent in B16 melanoma cells. Evidence revealed by an endoglycoce-ramidase. J Biol Chem 1996;271:12655–12660.

38. Ji L, Ito M, Zhang G, Yamagata T. The hydrolysis of cell sur-face glycosphingolipids by endoglycoceramidase reduces epidermalgrowth factor receptor phosphorylation in A431 cells. Glycobiology1995;5:343–350.

39. Tsai B, Gilbert JM, Stehle T, Lencer W, Benjamin TL, Rapoport TA.Gangliosides are receptors for murine polyoma virus and SV40. EMBOJ 2003;22:4346–4355.

40. Hanada K. Sphingolipids in infectious diseases. Jpn J Infect Dis2005;58:131–148.

41. Simons K, Ehehalt R. Cholesterol, lipid rafts, and disease. J Clin Invest2002;110:597–603.

42. Duncan MJ, Shin JS, Abraham SN. Microbial entry through caveolae:variations on a theme. Cell Microbiol 2002;4:783–791.

43. Lafont F, Abrami L, van der Goot FG. Bacterial subversion of lipidrafts. Curr Opin Microbiol 2004;7:4–10.

44. Richterova Z, Liebl D, Horak M, Palkova Z, Stokrova J, Hozak P,Korb J, Forstova J. Caveolae are involved in the trafficking of mousepolyomavirus virions and artificial VP1 pseudocapsids toward cellnuclei. J Virol 2001;75:10880–10891.

45. del Pozo MA, Balasubramanian N, Alderson NB, Kiosses WB, Grande-Garcia A, Anderson RG, Schwartz MA. Phospho-caveolin-1 mediatesintegrin-regulated membrane domain internalization. Nat Cell Biol2005;7:901–908.

46. Miyagi T, Wada T, Yamaguchi K, Hata K, Shiozaki K. Plasma mem-brane-associated sialidase as a crucial regulator of transmembranesignalling. J Biochem 2008;144:279–285.

47. Hakomori S. Tumor-associated carbohydrate antigens defining tumormalignancy: basis for development of anti-cancer vaccines. Adv ExpMed Biol 2001;491:369–402.

48. Hettmer S, Ladisch S, Kaucic K. Low complex ganglioside expres-sion characterizes human neuroblastoma cell lines. Cancer Lett2005;225:141–149.

49. Birkle S, Zeng G, Gao L, Yu RK, Aubry J. Role of tumor-associatedgangliosides in cancer progression. Biochimie 2003;85:455–463.

Traffic 2010; 11: 348–360 359

Singh et al.

50. Guerrera M, Ladisch S. N-butyldeoxynojirimycin inhibits murinemelanoma cell ganglioside metabolism and delays tumor onset. Can-cer Lett 2003;201:31–40.

51. Ling K, Doughman RL, Firestone AJ, Bunce MW, Anderson RA. TypeI gamma phosphatidylinositol phosphate kinase targets and regulatesfocal adhesions. Nature 2002;420:89–93.

52. Puri V, Jefferson JR, Singh RD, Wheatley CL, Marks DL, Pagano RE.Sphingolipid storage induces accumulation of intracellular choles-terol by stimulating SREBP-1 cleavage. J Biol Chem 2003;278:20961–20970.

53. Ledeen RW, Yu RK. Gangliosides: structure, isolation, and analysis.Methods Enzymol 1982;83:139–191.

360 Traffic 2010; 11: 348–360