PC12 cells have caveolae that contain TrkA. Caveolae-disrupting drugs inhibit NGF-, but not EGF-, induced MAPK phosphorylation

PC12 cells have caveolae that contain TrkA. Caveolae-disrupting drugs inhibit NGF-, but

not EGF-, induced MAPK phosphorylation.

Sandra Peiró1, Joan Comella2, Carlos Enrich1 Dionisio Martín-Zanca3 & Nativitat Rocamora1, 4

1Departament de Biologia Cel.lular i Anatomia Patològica, Institut d’Investigacions BiomèdiquesAugust Pi i Sunyer (IDIBAPS), Facultat de Medicina, Universitat de Barcelona. 2Grup deNeurobiologia Molecular, Departament de Ciències Mèdiques Bàsiques, Facultat de Medicina,Universitat de Lleïda.3Instituto de Microbiología Bioquímica, Consejo Superior de Investigaciones Científicas-Universidad de Salamanca.4Laboratori de Biologia Molecular, Institut Català d’Oncologia (ICO), l’Hospitalet de Llobregat.

Running title: Caveolae and NGF signaling in PC12 cells

Address for correspondence:

Nativitat RocamoraLaboratori de Biologia Molecular,Institut Català d’Oncologia (ICO).Avda Gran Via s/n. Km 2.7L’Hospitalet de Llobregat08907-Barcelonatel: 34-93-2607952fax: 34-93-2607741e-mail: [email protected]

Copyright 2000 by The American Society for Biochemistry and Molecular Biology, Inc.

JBC Papers in Press. Published on September 11, 2000 as Manuscript M000487200 by guest, on January 31, 2013

ww

w.jbc.org

Dow

nloaded from

http://www.jbc.org/

Caveolae and NGF signaling in PC12 cells

2

SUMMARY

Nerve Growth Factor (NGF) induces survival and differentiation of the neural-crest

derived PC12 cell line. Caveolae are cholesterol-enriched, caveolin-containing, plasma

membrane microdomains involved in vesicular transport and signal transduction. Here we

demonstrate the presence of caveolae in PC12 cells, and their involvement in NGF signaling.

Our results showed the expression of caveolin-1 by Western blot and confocal immuno-

microscopy. The presence of plasma membrane caveolae, was directly shown by rapid-freeze

deep-etching electron microscopy. Moreover, combined deep-etching and immunogold

techniques revealed the presence of the NGF receptor TrkA in caveolae of PC12 cells. These

data, together with the cofractionation of Shc, Ras, caveolin and TrkA, in the caveolae

fraction, supported a role for these plasma membrane microdomains in NGF signaling. To

approach this hypothesis, caveolae were disrupted by treatment of PC12 cells with

cholesterol-binding drugs. Either, filipin or cyclodextrin treatment increased basal levels of

MAPK-phosphorylation. In contrast, pretreatment of PC12 cells with these drugs inhibited the

NGF-, but not the EGF-, induced MAPK phosphorylation, without affecting the TrkA

autophosphorylation. Taken together, our results demonstrate the presence of caveolae in

PC12 cells which contain the high-affinity NGF receptor TrkA, and the specific involvement of

these cholesterol-enriched plasma membrane microdomains in the propagation of the NGF-

induced signal.

by guest, on January 31, 2013w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


3

INTRODUCTION

Caveolae are bottle shaped, plasma membrane invaginated pits, characterized by their size

(50-100 nm diameter), low-buoyant density, cholesterol enrichment, and striated coat of caveolin.

Although initially described by Palade (1) and Yamada (2) as small caves in the plasma membrane

of endothelial and epithelial cells, caveolae have been found in many other cell types, including

adipocytes, muscle cells, fibroblasts, astrocytes, pneumocytes and hepatocytes. The presence of

caveolin and the enrichment in cholesterol are the two main molecular hallmarks of caveolae. Thus,

80-90% of the plasma membrane cholesterol is concentrated in caveolae and contributes to the

characteristic low-buoyant density and detergent-insolubility of these plasma membrane

microdomains (3, 4). A family of caveolin proteins including caveolin-1, α and β forms, caveolin-2

and caveolin-3, has been described (5-7). Caveolins, are highly hydrophobic proteins with a

characteristic hairpin shape located inside the plasma membrane with both ends facing the

cytoplasm. The N-terminal domain of the protein mediates homo- and hetero-oligomerization of

the caveolin monomers as well as its binding to different molecules (8-11). Caveolae were though

to be mainly involved in transport of molecules across the cell. Thus, transcytosis and potocytosis

were described to be mediated by caveolae in endothelial cells (12-14). In fact, caveolae were

found to carry the molecular machinery required for vesicular transport (15, 16). However, the

presence of caveolin-associated preassembled signaling complexes, including membrane receptors

and downstream signal transducing molecules, supports a role for these plasma membrane

microdomains not only in transport but also in signal transduction (4, 17).

NGF is the prototypic member of the neurotrophin family of neurotrophic factors, which

also includes BDNF, NT3 and NT4/5 (18, 19). Two kinds of membrane receptor mediate

neurotrophin signaling: the Trk family of tyrosine kinase receptors, including TrkA, TrkB and TrkC


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


4

showing specific high affinity for NGF, BDNF/NT4 and NT3, respectively, and p75NTR, with

similar low affinity for all of them (20-22). Survival, differentiation, proliferation or cell death, may

be induced by neurotrophins, depending on the specific background and physiological state of a

cell. PC12 cells, derived from a pheochromatocytoma tumor of rat adrenal medulla (23), which

express both TrkA and p75NTR, have been extensively used to study the NGF action.

Undifferentiated, round-shaped PC12 cells grow exponentially in high serum conditions; in

response to NGF they stop growing, extend neurites and differentiate into cholinergic sympathetic-

like neurons (23). Moreover, NGF treatment could induce survival or cell death of serum-deprived

PC12 cells, depending on the relative amounts of TrkA and p75, respectively (24, 25).

In order to study the putative involvement of caveolae in NGF signaling, we have analyzed

the presence of these plasma membrane microdomains in PC12 cells. Our results showed the

expression of caveolin-1, together with the presence of prototypic caveolae clustered in specific

membrane subdomains. Moreover, the presence of the high-affinity NGF receptor TrkA in caveolae

supported their putative involvement in NGF signaling. Finally, we show that drug-induced

disruption of caveolae inhibits NGF-induced, but not EGF-induced, MAPK phosphorylation, with

no effect on the NGF-induced TrkA autophosphorylation. A role for caveolae in cellular

compartmentalization of signal transduction in PC12 cells is discussed.


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


5

EXPERIMENTAL PROCEDURES

Materials.Two polyclonal antibodies produced against TrkA were used: RTA raised

against the extracellular domain of rat TrkA (26) and a pan-trk antibody (α−203) raised against the

14 carboxy-terminal amino acids of human TrkA (27). Three antibodies raised against the N-

terminal domain of caveolin-1: a polyclonal (cavp) and two monoclonals, clones C060 and 2234,

and a polyclonal antibody against Shc, all of them from Transduction Laboratories (Lexington,

KY), were used. A monoclonal antibody, 4G10, directed against phosphorylated tyrosine and a

monoclonal antibody against pan-Ras were obtained from UBI (Lake Placid, NY).

Monoclonal antibodies against ERK-1 and ERK-2 were from Zymed Laboratories. Polyclonal

antibodies against phospho-ERK-1 and phospho-ERK-2 were from New England Biolabs

(Beverly, MA). TRITC-conjugated swine anti-rabbit and anti-mouse, fluorescein

isothiocyanate (FITC)-conjugated sheep anti-mouse and anti-rabbit were obtained from

Boehringer Mannheim Biochemicals. Peroxidase-conjugated swine anti-rabbit and rabbit anti-

mouse were obtained from BioRad. Goat anti-mouse conjugated to 10 or 15nm colloidal gold

and goat anti-rabbit IgG conjugated to 10 or 15nm colloidal gold were from BioCell. NGF

was purified from salivary glands as described previously (28). PC12 cells were obtained from

Dr. F. McKenzie (State University of New York, Stony Brook, NY).

Cell culture, cell treatments and cell lysates. PC12 cells were cultured in DMEM

supplemented with 5% fetal calf and 10% horse serums, 2 mM L-glutamine, 56 U/ml penicillin

and 56 µg/ml streptomycin. For cell treatments, PC12 cells, around 4 weeks after thawing,

were grown to 70% confluence and serum-deprived for 16-18h. Cells were then stimulated for

5 min with NGF (100ng/ml) or EGF (10ng/ml). For drug treatments PC12 cells were

incubated for 1h at 37º C in the presence of either vehicle alone or one of the following drugs:


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


6

10mM β-hydroxypropyl-Cyclodextrin, 5µg/ml Filipin or 5 µg/ml Xylazine (Sigma).

Reversibility of drug effects were achieved as follows: for filipin-treated cells, incubating with

normal media for 1 hour; for cyclodextrin-treated cells, with 16µg/ml cholesterol and 0.4%

cyclodextrin, in normal culture media, for 1 hour (29).

To obtain lysates, cells were quickly rinsed in ice-cold phosphate-buffered saline

(PBS), pH 7.2, and lysis was performed at 4ºC in 0.5 ml of buffer A (Tris 50mM, NaCl

150mM pH 7.4, 1 % Triton X-100 containing 1mM PMSF, 1mM aprotinin, 20µM leupeptin,

0.1 mM sodium vanadate, 1mM sodium fluoride and 10mM sodium pyrophosphate), or buffer

B (buffer A containing 60mM octylglucoside). After 45 minutes of incubation on ice, lysates

were centrifuged at 13,000 rpm in a microfuge at 4ºC for 30 minutes. Total cell extracts were

obtained with buffer C (67mM Tris pH 6.8 and 2% SDS). Protein was quantified by Bradford

or Lowry assay reagents (Bio Rad) and analyzed by SDS-PAGE (polyacrylamide gel

electrophoresis) and Western-blot.

Immunoprecipitation and Western blot. Cells were lysed in buffer B and 400-2000µg

protein in 1ml were used for immunoprecipitation studies. 4µg of rabbit anti-caveolin, or

mouse (clone 2234) anti-caveolin was used to immunoprecipitate caveolin. Pan-trk (α-203) or

anti-TrkA (RTA) antibodies were used to immunoprecipitate TrkA. Immunoprecipitation was

performed at 4ºC for 3h. Protein immunocomplexes were incubated with 20 µl of BSA-

blocked ProteinA-Sepharose (50% suspension in lysis buffer), for 1 hour at 4ºC, collected by

a short centrifugation (30 seconds), and washed five times in buffer B. Samples for

electrophoresis were solubilized with Laemmli sample buffer containing 10% 2-

mercaptoethanol, heated to 95ºC for 5 min and resolved by SDS-PAGE using either 10% or

12.5% acrylamide. For Western blot, proteins were transferred to immobilon-P strips for 2 h


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


7

at 60 V, blocked with 5% dried milk (except for the anti-phosphotyrosine antibody 4G10

which was blocked with 5% BSA) in TBS and incubated for 1 hour at room temperature with

primary antibodies (anti-caveolin-1 polyclonal 1:3000; RTA 1:500; anti- ERK-1 and ERK-2

1:500; anti- phospho-ERK-1 and ERK-2 1:500; anti-phosphorylated tyrosine 4G10 1:1000;

anti-Shc 1:500; anti-pan-Ras 1:100) diluted in blocking solution. After three washes in TBST

containing 0.05% Tween 20, the membrane was incubated with appropriate secondary

antibodies conjugated with peroxidase, diluted 1:2000 in blocking solution. Developing was

performed with the ECL kit.

Immunostaining and confocal microscopy. PC12 cells were plated on tissue culture

dishes containing sterile glass coverslips coated with collagen and poly-ornitin. Cells were

washed twice in PBS, fixed for 30 min with 3% paraformaldehyde in PBS buffer pH 7.4 and

permeabilized with 0.1 % Triton X-100. Cells were then blocked with fetal calf serum (1%) in

PBS-Gly for 30 min and incubated with the first antibody diluted in blocking solution.

Dilutions used were 1:150 for anti-cav (polyclonal or monoclonal), 1:500 for anti-TrkA (RTA

or α−203). The incubation was carried out in a humidified chamber, at 37 ºC for 45 min.

Three washes in PBS were performed to remove excess antibody before adding the secondary

fluorescent-conjugated antibody (FITC-conjugated sheep anti mouse or TRITC-conjugated

swine anti-rabbit). Coverslips were mounted on glass slides using Mowiol. Confocal images

were collected using a Leica TCS NT equipped with a 63X Leitz Plan-Apo objective (NA

1.4). Adobe Photoshop software (Adobe Systems, San Jose, CA) was used for image

processing.


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


8

Detergent-free, cell-membrane fractionation. PC12 cells grown to confluence in 150-

mm diameter dishes were used to obtain caveolin-enriched membrane fractions essentially as

described (30). Cyclodextrin treated or untreated cells were used. After two washes in ice-

cold PBS, PC12 cells from a confluent 150-mm flask were scraped in 0.5 ml of 500 mM

sodium carbonate, pH 11.0, transferred to a plastic tub and homogenized in two steps; using a

0.9 x 40 needle and a sonicaton bath (5 min.). The homogenate was adjusted to 45% sucrose

by the addition of 0.5 ml of 90% sucrose prepared in MBS (25 mM MES, pH 6.5, 0.15 M

NaCl) and placed at the bottom of an ultracentrifuge tube. A 5-35% discontinuous sucrose

gradient (w/v) was formed above (0.6 ml of 5% sucrose, 2.4 ml of 35% sucrose; both in MBS

containing 250 mM sodium carbonate) and centrifuged at 36,000 rpm for 7 h at 4ºC in an

TI50 rotor. Sucrose gradient fractions (450µl) were collected from the top of the

ultracentrifuge tubes and analyzed by SDS-PAGE

Electron microscopy. Rapid freezing and freeze-substitution. PC12 cells were grown

on transwell filters in order to facilitate further sectioning. Small pieces of the filters were

cryofixed by projection against a copper block cooled in liquid nitrogen (-196ºC) using a

Cryoblock (Leica) as described (31). Freeze-substitution was performed in a home-made

cryosystem (32), using acetone containing 0.5% of uranyl acetate, for three days at –90ºC. On

day 4 the temperature was slowly increased, 5ºC/h, to –50ºC. At this temperature samples

were rinsed in acetone and then infiltrated and embedded in Lowicryl HM20, as follows: 50%

Lowicryl in acetone for 4 hours; 75% Lowicryl in acetone overnight; 100% Lowicryl

overnight and 100% fresh resin for 4 hours. After infiltration resin blocks were prepared and

polymerized using UV light. Ultrathin sections were picked up on Formvar-coated gold grids

in order to carry out the immunocytochemical labeling: Samples were blocked with 2%


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


9

ovoalbumin for 30 minutes and incubated at room temperature for 1 hour with rabbit anti-pan-

trk antibody (α-203, 1:500) and mouse anti-caveolin-1 antiboby (C060, 1:150). Washes were

performed to remove excess antibody with PBS prior to adding goat anti-mouse or anti-rabbit

IgG conjugated to 15nm and 10nm colloidal gold, respectively, for 45 minutes at room

temperature. Finally, samples were washed and contrasted with 2% uranyl acetate.

Rapid-freezing and deep-etching. PC12 cells were grown to confluence on tissue

culture dishes containing 8-12 sterile glass coverslips coated with collagen and poly-ornitin.

Upper plasma membrane was removed by sonication (1 second burst) in an isotonic buffer that

contained 70mM KCl, 3mM EGTA, 30mM HEPES and 5mM MgCl2 after a hypotonic shock

in 1/3 of isotonic buffer for 1 minute. Coverslips were immediately fixed for 30 minutes at

room temperature with 3% paraformaldehyde in cold PBS pH 7.4, then blocked with 10%

fetal calf serum for 30 min and incubated with rabbit pan-trk (1:500) and mouse anti-caveolin-

1 (1:150) antibodies. The incubation was carried out at room temperature for 45 min. Three

washes were performed to remove excess antibody prior to adding goat anti-mouse or anti-

rabbit IgG conjugated to 10nm and 15nm colloidal gold, respectively, for 30 min at room

temperature. Samples were fixed with 2.5 % glutaraldehyde before processing as indicated

below for rapid-freezing and deep-etching electron microscopy. Briefly, coverslips were cut

into 2-3 mm squares and cryofixed by projection against a copper block cooled in liquid

nitrogen (-196ºC) using a Cryoblock (Reichert-Jung, Leica). The frozen samples were stored

at –70ºC until subsequent use. Samples were etched and coated with platinum and carbon

using a freeze-etching unit (model BAF-060, BAL-TEC; Liechtenstein.). A rotatory replica of

the exposed surface was prepared by evaporating 1 nm platinum-carbon at an angle of 24º

above the horizontal, followed by 10 nm of carbon evaporated at a 75º angle (33). The replica


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


10

was separated from the coverslip by immersion in full-strength hydrofluoric acid, washed

twice in distilled water and digested with 5% sodium hypochlorite for 5-10 minutes. Finally,

the replicas were washed several times in distilled water, broken into small pieces, and picked

up on Formvar-coated copper grids for electron microscopy. When immunolabeling was

carried out the replicas were only washed in distilled water. All electron micrographs were

obtained using an electron microscope Hitachi HU-600, operating at 75 KV.

The photographic negatives were digitalized without contrast reversing and treated by the

IMAT program (Alejandro DiGiorgio, Serveis Científico Tècnics, Universitat de Barcelona).


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


11

RESULTS

Wild-type PC12 cells express caveolin and have caveolae. Confocal microscopy analyses

were performed, in paraformaldehyde-fixed PC12 cells, with a monoclonal and a polyclonal

antibody, both raised against the N-terminal domain of caveolin-1 (Fig.1A). The polyclonal

antibody has been shown to recognize all known members of the caveolin family, whereas the

monoclonal (C060) recognizes only α and β isoforms of caveolin-1. Our immunofluorescence

results showed different patterns of staining with the two antibodies: the polyclonal anti-

caveolin-1 (a) revealed a general and punctate distribution of caveolin in asynchronic

proliferating wild-type PC12 cells, whereas the monoclonal (b) showed a more restricted

plasma membrane distribution of the caveolin immunoreactivity.

Western blot analysis with the polyclonal anti-caveolin-1 antibody revealed a major

protein band with an apparent electrophoretic mobility of around 25kDa in total cell extracts

of Swiss3T3, NIH3T3 and NRK cells. In contrast, none of the two monoclonal antibodies

against caveolin-1, 2234 and C060, revealed any immunoreactive band when used in Western

blot (not shown).

Because the characteristic Triton X-100 insolubility of the oligomeric caveolin (10),

two lysis buffers containing, either Triton X-100 or Triton X-100 and octylglucoside, were

used. Caveolin-immunoreactivity was partially solubilized (around 50%) by 1% Triton X-100

and almost all solubilized by a lysis buffer containing 1% Triton X-100 plus 60mM

octylglucoside, supporting the presence of two different pools of caveolin in PC12 cells.

Other, higher molecular weight caveolin-1 immunoreactive bands, probably revealing SDS-

resistant caveolin-oligomers and/or other caveolin-like proteins, were also observed when

films were overexposed (not shown).


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


12

Specificity of the 25kDa band was assessed by combining immunoprecipitation and

Western blot. Immunoprecipitation with anti-caveolin-1, either monoclonal (2234) or

polyclonal, antibodies followed by Western blot with the polyclonal revealed only the 25kDa

band (Fig.1B).

Rapid-freeze deep-etching electron microscopy, which revealed the internal face of the

basal plasma membrane, was performed, in PC12 cells. Electron microscopy analysis showed

different plasma membrane invaginations: clathrin-coated pits and caveolae. Although relatively

smaller, between 100-200nm in diameter, than the prototypic clathrin-coated pits, the characteristic

basket-like structure of the clathrin coat was clearly seen (Fig. 2A). Other smaller invaginations,

around 50-100nm in diameter, which showed the characteristic striated-coat of caveolae were also

observed (Fig.2B). Combined rapid-freeze deep-etching and immuno-gold techniques using the

polyclonal anti-caveolin-1 antibody revealed the presence of caveolin in caveolae-like membrane

invaginations and also outside the caveolae (Fig.2C). Caveolin-immunoreactivity, however, was

completely absent from the clathrin-coated-pit enriched areas. Moreover, clathrin-coated pits and

prototypic caveolae were not interspersed in the plasma membrane, but mainly distributed in

different plasma membrane membrane subdomains (Fig.2 A & B).

TrkA, caveolin and caveolae. Double immunofluorescence was performed with polyclonal

pan-Trk (α−203) or the TrkA specific RTA antibodies, and a monoclonal anti-caveolin-1 antibody

(C060), on PC12 cells grown on collagen and polyornitine-coated coverslides. Confocal

microscopy analysis showed colocalization of TrkA and caveolin at the plasma membrane in

proliferating PC12 cells (Fig.3A).


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


13

Double immunogold with the polyclonal anti-pan-trk antibody (α-203) and the

monoclonal anti-caveolin-1 (C060), were performed on Lowicryl-included ultrathin sections of

exponentially growing PC12 cells and in rapid-freeze deep-etched material. Electron microscopic

data showed the presence of TrkA in prototypic caveolae (Fig.3B). Moreover, postembedding

double immunogold labeling of Lowicryl included material showed typical bottle-shaped membrane

invaginations of 50-100nm in diameter containing both TrkA (10nm gold particle), and caveolin-1

(15nm gold particle) immunoreactivity. In contrast, the electron microscopy data did not show

TrkA immunoreactivity in clatrhin-enriched plasma membrane domains (Fig.3B).

Cholesterol-depletion of PC12 cells. PC12 cells were treated with cholesterol-binding

drugs. Confocal microscopy showed that treatment with cyclodextrin (known to remove

cholesterol), in contrast to filipin or xylazine, increased caveolin immunoreactivity in the

perinuclear area of PC12 cells (Fig.4A). Detergent-free, cell membrane fractionation of

cyclodextrin-treated or untreated cells was performed. Western blot analysis showed TrkA,

caveolin, Shc and Ras cofractioning in the upper fractions of the sucrose gradient, interfase

between 5-35% sucrose (fractions 2 and 3), in untreated PC12 cells. Cyclodextrin-treatment,

and thus cholesterol-depletion of the cells, caused a shift to higher bouyant-density of

membranes containing TrkA, caveolin, Shc or Ras proteins (Fig.4B).

Involvement of caveolae in NGF signaling. Stimulation of PC12 cells with NGF

(100ng/ml) for 5 min induced a significant increase of MAPK phosphorylation. Two

cholesterol-binding drugs known to functionally disrupt caveolae, filipin and cyclodextrin,

were used to evaluate the putative involvement of cholesterol-sensitive structures, such as

caveolae, in NGF signaling. Xylazine, a general lipid-binding drug, which does not show

specific affinity for cholesterol, was used as a control. Filipin and cyclodextrin slightly


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


14

increased basal MAPK phosphorylation levels in PC12 cells (not shown, only detected in

overexposed films). However, pretreatment of PC12 cells with either cyclodextrin or filipin,

but not with xylazine, inhibited NGF-induced MAPK phosphorylation. Levels of solubilized

MAPK in cell lysates from drug-treated or untreated cells were not altered. Drug effects were

reversed incubating PC12 cells with cyclodextrin-cholesterol (for cyclodextrin-treated cells) or

with normal media (for filipin) (Fig. 5A). NGF-induced phosphorylation of TrkA was

unaltered by any of the cholesterol-binding drugs treatments (Fig. 5B), respect to the control

(Xylazin). In contrast to NGF, the EGF-induced MAPK phosphorylation was unmodified by

pretreatment with any of the drugs (Fig. 5C).


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


15

DISCUSSION

The aims of the present study were to show the presence of caveolae, in PC12 cells, and to

analyze its putative involvement in NGF signaling. Both, the expression of caveolin-1 and the

presence of plasma membrane caveolae microdomains have been showed in wild-type proliferating

PC12 cells. A putative involvement of caveolae in NGF signaling was supported by the presence of

TrkA, the high-affinity NGF receptor, together with other signaling intermediaries such as Shc and

Ras in caveolae. Functional experiments, in which the integrity of caveolae was disrupted by

pretreatment with the cholesterol-binding drugs filipin or cyclodextrin, showed the specific

involvement of these cholesterol-enriched plasma membrane microdomains in the propagation of

NGF-induced signal. Taken together, our data support a role for caveolae in specific

compartmentalization of signal transduction in PC12 cells.

Three methodological aspects have to be taken into account for the discussion of the results

presented here: (i) the use of wild-type PC12 cells (ii) the direct detection of caveolae by electron

microscopy, combining rapid-freeze, deep-etching or Lowicryl thin sections with immunogold

techniques and (iii) the “in vivo” approach, to study the putative involvement of caveolae in NGF

signaling.

The expression of caveolin in PC12 cells is controversial. The absence of caveolin was

specifically reported for these cells (34). However, caveolin-immunoreactivity has been

actually detected in undifferentiated PC12 cells (35, 36). Moreover, both, expression of

caveolin-1 and upregulation of caveolin-2 mRNAs have been described, in differentiating and

stressed PC12 cells, respectively (35). The above differences could be, at least, partially

attributed to the use of various antibodies (36, and present results). We showed the expression

of caveolin-1, in wild-type proliferating PC12 cells, by both immunofluorescence and Western


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


16

blot analyses. However, in our hands, only the polyclonal antibody showed caveolin-

immunoreactivity by Western blot. Moreover, the different immunofluorescence patterns

obtained with the polyclonal or monoclonal antibodies, together with the presence of various

caveolin-immunoreactive protein bands in Western blot overexposed-films, supported the

possibility of other members of the caveolin family and/or caveolin-like proteins expressed in

these cells, as suggested by Cameron et al. (37).

Direct evidence for the presence of caveolae has not been shown either in wild-type or

in caveolin-overexpressing PC12 cells. The combined deep-etching and immunogold

techniques, of the present study, allowed us to detect both caveolae and clathrin-coated pits in

wild-type PC12 cells. These results added further evidence of the widespread presence of

caveolae and suggest a role for these plasma membrane microdomains in cell physiology. In

addition, a specific enrichment of caveolae or clathrin-coated pits was found at different

membrane subdomains of the proliferating PC12 cells, suggesting the presence of preformed

plasma membrane specializations in these undifferentiated cells.

Because caveolin is the only proteinic hallmark of caveolae, colocalization of TrkA and

caveolin-1 suggested the presence of the high-affinity NGF receptor in caveolae of PC12 cells.

However, caveolin was not only restricted to caveolae, as supported by, its partial solubility in

Triton X-100, and the deep-etching-immunoglod data. In the present study, combination of

Lowicryl or rapid-freeze deep-etching with double-immunogold detection of TrkA and

caveolin, gave a direct evidence for the presence of TrkA in caveolae of PC12 cells.

Caveolae-enrichment was obtained by a detergent-free sucrose-gradient fractionation

method. However, cofractionation with caveolin, by this method, cannot be considered by

itself as an evidence for the presence of a given molecule in caveolae, as extensively discussed


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


17

in a recent paper (38). The EGF receptor is a case in point. Although it cofractionated with

caveolin in A341 cells, it did not coimmunoprecipitate in immunoabsortion of the caveolin-enriched

fraction with the monoclonal anti-caveolin-1 2234 antibody (39). Comparison of the detergent-

free sucrose-gradient fractionation obtained from cyclodextrin-treated or untreated cells

combined two criteria for the assignment of a given molecule to caveolae: the low-bouyant

density and the cholesterol enrichment of these plasma membrane microdomains. Using this

approach, we showed cofractionation of TrkA, caveolin, Shc and Ras in low-bouyant density

membranes, together with their shift to higher densities after cyclodextrin-treatment, and thus

membrane depletion of cholesterol, in PC12 cells. The presence of TrkA and caveolin, together

with other signaling intermediaries, such as Shc and Ras, in caveolae supported a putative

involvement of caveolae in NGF signaling.

Wild-type PC12 cells were pretreated with cholesterol-binding drugs, either filipin or

cyclodextrin, known to disrupt caveolae and interfere with their functionality (14, 40, 41), and

the effect on NGF signaling was analyzed. Although both drugs bind cholesterol only

cyclodextrin removes it from the plasma membrane (42). Cyclodextrin treatment of PC12 cells

induced accumulation of caveolin in the perinuclear area. These results, together with the

described effect of cholesterol-oxidase (another cholesterol-removing drug) on the

traslocation of caveolin to the Golgi (43), supported a role for cholesterol in shuttling caveolin

to the plasma membrane.

Caveolae involvement in signaling was mainly based on a caveolin scaffolding function,

which mediates the accumulation of preassembled signaling complexes at caveolae. A general

inhibitory role for caveolin has been supported by its specific ability to bind inactive forms of

signaling molecules (30, 44, 45). Such a role has been specially well documented for the NOS


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


18

system (46-49). In contrast, caveolin has been found positively involved in the integrin-

mediated activation of Fyn (50), and in the activation of the insulin receptor signaling (51).

The association of TrkA and caveolin together with an inhibitory effect of caveolin on TrkA

autophosphorylation, have recently been described in caveolin/TrkA-overexpressing PC12

cells (36).

In agreement with Furuchi and Anderson (29), a drug-induced increase in MAPK

phosphorylation was observed, showing an inhibitory effect of intact caveolae on basal MAPK

phosphorylation. In contrast, functional-disruption of caveolae by pretreatment of PC12 cells

with either cyclodextrin or filipin, inhibited the NGF-induced MAPK phosphorylation, with no

effect on TrkA autophosphorylation. Thus, a dual effect for caveolae was shown: an inhibitory

effect on basal MAPK-phosphorylation, together with its requirement for the propagation of

the NGF-induced signal. Cholesterol repletion of PC12 cells which have been pretreated with

caveolae-disrupting, cholesterol-binding drugs, restored their ability to activate MAPK upon

NGF stimulation. These results further supported a role for intact caveolae or cholesterol-

enriched membrane microdomains in NGF signaling. In contrast to NGF, the EGF signal

transduction, at least at the level of MAPK phosphorylation, was unaffected by any of the

drugs used, showing their specificity, in our experimental conditions. A role for caveolae in

specific compartmentalization of signaling was supported by the present results.

The PC12 cell line is a good model to perform comparative studies of signaling

pathways triggering cell differentiation and proliferation. Whereas these cells proliferate in

response to high serum or EGF, they stop growing and differentiate into cholinergic-like

neurons, in response to NGF. The specificity of a particular signaling pathway is provided

initially by the particular ligand/receptor system involved; however, there may be a


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


19

convergence downstream, where various signal transducing molecules are shared. Not only

the signal transducing molecules involved, but their macromolecular organization, subcellular

trafficking and localization could be important for specific regulation of signaling. Our data

showed TrkA preferentially associated with caveolae and caveola-enriched areas of the plasma

membrane whereas it was essentially depleted from clathrin-enriched domains, in unstimulated

PC12 cells. In addition, the functional integrity of these cholesterol-enriched domains was

shown to be necessary for the propagation of the NGF induced signal, at least up to the level

of MAPK-phosphorylation. In contrast, activated TrkA bound to PLC-γ was found in

internalized clathrin-immunoreactive vesicles, after 20 min estimulation with NGF (52, 53). An

NGF-induced translocation of TrkA from caveolae, in unstimulated PC12 cells, to other

cellular compartments, such as intracellular clathrin-containing organelles, could be suggested.

Dynamic cellular microdomains may play an important role in spatio-temporal regulation of

signal transduction.


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


20

REFERENCE LIST

1. Palade GE (1953) J.Appl.Physics 24, 1424

2. Yamada E (1955) J.Biophys.Biochem.Cytol 1, 445-458

3. Parton, R.G. (1996) Curr.Opin.Cell Biol. 8, 542-548

4. Anderson, R.G. (1998) Annu.Rev.Biochem 67: 199-225

5. Glenney, J.R.J. and Soppet, D. (1992) Proc.Natl.Acad.Sci.U.S.A. 89, 10517-10521

6. Scherer, P.E., Okamoto, T., Chun, M., Nishimoto, I., Lodish, H.F., and Lisanti, M.P.(1996) Proc.Natl.Acad.Sci.U.S.A. 93, 131-135

7. Tang, Z., Scherer, P.E., Okamoto, T., Song, K., Chu, C., Kohtz, D.S., Nishimoto, I.,Lodish, H.F., and Lisanti, M.P. (1996) J.Biol.Chem. 271, 2255-2261

8. Lisanti, M.P., Tang, Z.L., and Sargiacomo, M. (1993) J.Cell Biol. 123, 595-604

9. Sargiacomo, M., Scherer, P.E., Tang, Z., Kubler, E., Song, K.S., Sanders, M.C., andLisanti, M.P. (1995) Proc.Natl.Acad.Sci.U.S.A. 92, 9407-9411

10. Monier, S., Parton, R.G., Vogel, F., Behlke, J., Henske, A., and Kurzchalia, T.V. (1995) Mol.Biol.Cell 6, 911-927

11. Monier, S., Dietzen, D.J., Hastings, W.R., Lublin, D.M., and Kurzchalia, T.V. (1996) FEBS Lett. 388, 143-149

12. Anderson, R.G., Kamen, B.A., Rothberg, K.G., and Lacey, S.W. (1992) Science 255,410-411

13. Parton, R.G., Joggerst, B., and Simons, K. (1994) J.Cell Biol. 127, 1199-1215

14. Schnitzer, J.E., Oh, P., Pinney, E., and Allard, J. (1994) J.Cell Biol. 127, 1217-1232

15. Schnitzer, J.E., Oh, P., and McIntosh, D.P. (1996) Science 274, 239-242

16. Scheiffele, P., Verkade, P., Fra, A.M., Virta, H., Simons, K., and Ikonen, E. (1998) JCell Biol 140, 795-806

17. Okamoto, T., Schlegel, A., Scherer, P.E., and Lisanti, M.P. (1998) J Biol Chem. 273 ,5419-5422

18. Levi-Montalcini, R. (1987) EMBO J. 6, 1145-1154

19. Lewin, G.R. and Barde, Y.A. (1996) Annu.Rev.Neurosci. 19: 289-317


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


21

20. Barbacid, M. (1995) Curr.Opin.Cell Biol. 7, 148-155

21. Chao, M., Casaccia-Bonnefil, P., Carter, B., Chittka, A., Kong, H., and Yoon, S.O.(1998) Brain Res Brain Res Rev. 26, 295-301

22. Frade, J.M. and Barde, Y.A. (1998) Bioessays 20, 137-145

23. Greene, L.A. and Tischler, A.S. (1976) Proc.Natl.Acad.Sci U.S.A. 73, 2424-2428

24. Barrett, G.L. and Georgiou, A. (1996) J Neurosci.Res 45, 117-128

25. Taglialatela, G., Robinson, R., and Perez-Polo, J.R. (1997) J Neurosci.Res 47, 155-162

26. Clary, D.O., Weskamp, G., Austin, L.R., and Reichardt, L.F. (1994) Mol.Biol.Cell 5,549-563

27. Martin-Zanca, D., Oskam, R., Mitra, G., Copeland, T., and Barbacid, M. (1989) Mol.Cell Biol 9, 24-33

28. Mobley, W.C., Schenker, A., and Shooter, E.M. (1976) Biochemistry 15, 5543-5552

29. Furuchi, T. and Anderson, R.G. (1998) J Biol Chem. 273, 21099-21104

30. Song, S.K., Li, S., Okamoto, T., Quilliam, L.A., Sargiacomo, M., and Lisanti, M.P.(1996) J.Biol.Chem. 271, 9690-9697

31. Escaig J (1982) J.Microscopy 126, 221-229

32. Quintana, C. (1994) Micron. 25, 63-99

33. Heuser, J.E. and Anderson, R.G. (1989) J Cell Biol 108, 389-400

34. Bilderback, T.R., Grigsby, R.J., and Dobrowsky, R.T. (1997) J Biol Chem. 272, 10922-10927

35. Galbiati, F., Volonte, D., Gil, O., Zanazzi, G., Salzer, J.L., Sargiacomo, M., Scherer,P.E., Engelman, J.A., Schlegel, A., Parenti, M., Okamoto, T., and Lisanti, M.P.(1998) Proc.Natl.Acad.Sci U.S.A. 95, 10257-10262

36. Bilderback, T.R., Gazula, V.R., Lisanti, M.P., and Dobrowsky, R.T. (1999) J BiolChem. 274, 257-263

37. Cameron, P.L., Ruffin, J.W., Bollag, R., Rasmussen, H., and Cameron, R.S. (1997) JNeurosci. 17, 9520-9535

38. Oh, P. and Schnitzer, J.E. (1999) J.Biol.Chem. 274, 23144-23154

39. Waugh, M.G., Lawson, D., and Hsuan, J.J. (1999) Biochem J 337, 591-597


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


22

40. Chang, W.J., Rothberg, K.G., Kamen, B.A., and Anderson, R.G. (1992) J.Cell Biol.118, 63-69

41. Baorto, D.M., Gao, Z., Malaviya, R., Dustin, M.L., van der Merwe, A., Lublin, D.M.,and Abraham, S.N. (1997) Nature 389, 636-639

42. Neufeld, E.B., Cooney, A.M., Pitha, J., Dawidowicz, E.A., Dwyer, N.K., Pentchev,P.G., and Blanchette-Mackie, E.J. (1996) J.Biol.Chem. 271, 21604-21613

43. Smart, E.J., Ying, Y.S., Conrad, P.A., and Anderson, R.G. (1994) J.Cell Biol. 127,1185-1197

44. Li, S., Couet, J., and Lisanti, M.P. (1996) J.Biol.Chem. 271, 29182-29190

45. Engelman, J.A., Chu, C., Lin, A., Jo, H., Ikezu, T., Okamoto, T., Kohtz, D.S., andLisanti, M.P. (1998) FEBS Lett. 428, 205-211

46. Garcia-Cardena, G., Fan, R., Stern, D.F., Liu, J., and Sessa, W.C. (1996) J.Biol.Chem.271, 27237-27240

47. Garcia-Cardena, G., Martasek, P., Masters, B.S., Skidd, P.M., Couet, J., Li, S., Lisanti,M.P., and Sessa, W.C. (1997) J Biol Chem. 272, 25437-25440

48. Michel, J.B., Feron, O., Sacks, D., and Michel, T. (1997) J Biol Chem. 272, 15583-15586

49. Ju, H., Zou, R., Venema, V.J., and Venema, R.C. (1997) J Biol Chem. 272, 18522-18525

50. Wary, K.K., Mariotti, A., Zurzolo, C., and Giancotti, F.G. (1998) Cell 94, 625-634

51. Yamamoto, M., Toya, Y., Schwencke, C., Lisanti, M.P., Myers, M.G.J., and Ishikawa,Y. (1998) J Biol Chem. 273, 26962-26968

52. Grimes, M.L., Zhou, J., Beattie, E.C., Yuen, E.C., Hall, D.E., Valletta, J.S., Topp, K.S.,LaVail, J.H., Bunnett, N.W., and Mobley, W.C. (1996) J Neurosci. 16, 7950-7964

53. Grimes, M.L., Beattie, E., and Mobley, W.C. (1997) Proc.Natl.Acad.Sci U.S.A. 94,9909-9914


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


23

FOOTNOTES

1 The abbreviations used are: NGF, nerve growth factor; BDNF, brain-derived neurotrophic

factor; NT, neurotrophin; ERK, extracellular-regulated kinase; MAPK, mitogen-activated

protein kinase; Trk, tyrosine kinase receptor; NTR, neurotrophin receptor.

ACKNOWLEDGEMENTS

We are very grateful to Neus Agell for helpful discussions and critical reading of the

manuscript, and to Anna Bosch and Carmen López-Iglesias for technical assistance in confocal and

electron microscopy, respectively. We thank Robin Rycroft for editorial help. The present study

was supported by grants PM96/0083 to C. E. and Marató TV3/98 to J.C. D.M.-Z. was supported

by grants from the Spanish Ministry for Education and Culture PB94/1104 and CE B104-CT96-

0285. S.P. is a recipient of a predoctoral fellowship from the IDIBAPS foundation.


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


24

FIGURE LEGENDS

Figure 1.- Wild-type PC12 cells express caveolin-1

A. Confocal microscopy showing caveolin-1 expression in proliferating PC12 cells.

Immunocytochemistry was performed with polyclonal (a) and monoclonal (b) anti-caveolin-1

antibodies. Note the general and diffuse signal distribution revealed by the polyclonal antibody in

contrast to the more membrane-restricted pattern shown by the monoclonal antibody. Bar 5µm. B.

Western blot analysis of caveolin expression with the polyclonal anti-caveolin-1 antibody. Left

panel shows total cell lysates from Swiss3T3, NIH3T3 and NRK cells, obtained with the SDS-

containing buffer (see Material and Methods). Panel in the middle shows different soluble (S) and

insoluble (P) pools of caveolin-1, obtained from PC12 cells using Triton X-100 (S1/P1) or

octylglucoside (S2/P2) containing buffers (see Material and Methods). Panel on the right, shows

Western blot analyses using the polyclonal anti-caveolin-1 antibody (cavp), after

immunoprecipitation (Ip) with the same antibody or the monoclonal 2234 antibody.

Figure 2.- PC12 cells have plasma membrane prototypic caveolae

Rapid-freeze deep-etching electron microscopy showing separated plasma membrane domains

differentially enriched in clathrin or caveolae-like vesicles. Note in panel A, the presence of

prototypic large clathrin-coated membrane domains, as invaginated pits (arrow) or showing a flat

appearance (arrowhead). Panel B shows prototypic caveolae vesicles characterized by their size

(50-100nm) and striated appearance (arrows).


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


25

In panel C, combination of the deep-etching and immuno-gold techniques with the polyclonal anti-

caveolin-1 antibody. Caveolae show the characteristic size (50-100nm) and the presence of

caveolin (arrows). The lower resolution in panel C is due to the combination of immuno-gold and

deep-etching methods. Note the caveolin immunoreactivity in caveolae (arrows) and also outside

the caveolae (arrowheads). Bars 100nm.

Figure 3.- Presence of TrkA in caveolae of PC12 cells

A. Confocal microscopy of round-shaped proliferating PC12 cells, showing the distribution, in

green, of the high-affinity NGF receptor TrkA, revealed by the polyclonal pan-trk antibody

(α−203), of caveolin-1, revealed by the monoclonal anti-caveolin-1 (C060) in red, and the

colocalization of both, in yellow. Note, the high degree of colocalization of TrkA and

caveolin-1 in the membrane area. Bar is 5µm. B. Electron microscopy analysis of the presence

of TrkA in prototypic caveolae. In a, postembedding double immunocytochemistry with the

polyclonal pan-trk (α-203) and the monoclonal (C060) anti-caveolin-1, in Lowicryl ultrathin

sections of PC12 cells. Note an invaginated membrane vesicle of about 50nm of diameter

showing both gold particles of 10 and 15nm corresponding to Trk (arrowhead) and caveolin

(arrow) immunoreactivities, respectively. In b and c, double immunocytochemistry in rapid-

freeze deep-etched material, with antibodies α-203 and C060 and secondaries conjugated with

gold particles of 15 and 10nm, respectively. Note the presence, in b, of vesicles containing

both TrkA and caveolin immunoreactivities (arrows) and the absence of both from the

clathrin-coated pits, in c. Bar in a is 50nm, and in b and c 100nm.


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


26

Figure 4.- Cholesterol-binding drugs treatment of PC12 cells

PC12 cells grown on collagen/polyornitine-coated coverslips and treated, for 1h at 37ºC, with

cholesterol-binding drugs: cyclodextrin or filipin, or with a general lipid-binding drug xylazine. A.

Immunocytochemistry against caveolin was performed using the polyclonal anti-caveolin-1

antibody and a FITC-conjugated secondary antibody. The four panels in A, are confocal

photomicrographs showing caveolin distribution in drug-untreated control PC12 cells (a), or after

treatment with: xylazine (b), cyclodextrin (c) or filipin (d). Note the accumulation of caveolin at the

perinuclear region after cyclodextrin treatment. Bar is 10µm. B. Cofractionation of TrkA, caveolin,

Shc and Ras. SDS-PAGE analysis of fractions, 1 to 6, from the Na-carbonate detergent-free

fractionation method followed by Western blot analyses. Note the cofractionation of both TrkA

110 and 140kDa forms and caveolin in fractions 2,3 (corresponding to 20% sucrose ) of the

control cells gradient, and its shift together to heavier fractions in cyclodextrin treated cells.

Figure 5.- Intact caveolae are required for NGF signal propagation

Western blot analysis showing MAPK-phosphorylation (A and C) and TrkA phosphorylation

(B) of PC12 after different treatments. Two cholesterol-binding drugs, cyclodextrin and filipin,

were used to functionally disrupt caveolae. Xylazine was used as a control. PC12 cells were

grown up to 70% confluence, serum-deprived for 16 hours, and incubated with cyclodextrin

(CD), filipin (F), xylazine (Xy), or with vehicle alone (C), for 1h at 37ºC. Drugs treatment

effects were reversed with normal medium (F+m) for filipin, or cholesterol (CD+col) for

cyclodextrin (Figure 5A). After treatments cells were stimulated or not for 5 min with NGF

(Figure 5A, B) or with EGF (Figure 5C). MAPK phosphorylation levels in A and C were

detected by western blotting of total cell extracts using a polyclonal anti-phospho-ERK-1 and


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/


27

phospho-ERK-2. Total MAPK levels were obtained by reprobing the corresponding blots with

monoclonal anti-ERK-1 and ERK-2 antibodies. Note that cyclodextrin and filipin, but not

xylazine, inhibited the NGF-induced MAPK-phosphorylation; and that the NGF ability to

activate MAPK was recovered by re-incubation of drug-treated cells with cholesterol (for

cyclodextrin) or normal media (for filipin) (A). The EGF-induced MAPK-phosphorylation was

not inhibited by any of the drugs used (C). NGF-induced phosphorylation of TrkA after

different treatments (panel B). SDS-PAGE of RTA-immunoprecipitated TrkA followed by

Western blot with the phospho-tyrosine antibody 4G10 and rebloting with the pan-Trk

antibody (α-203), after different pretreatments. Note a slightly decrease in TrkA

phosphorylation after the three drug-treatments, respect to untreated control cells. However,

TrkA phosphorylation levels were unaltered by any of the two cholesterol-binding drugs

cyclodextrin or filipin, respect to xylazine (unspecific lipid-binding drug, used as a control)


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

A B

Swiss NIH NRK S1 P1 S2 P2 Ip: cavp 2234

PC12

31-

21-

45-

31-

21-

45-

31-

21-

45-

31-

21-

45-

a

b

figure 1


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

A

B

C

figure 2


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

A

B

figure 3


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

A

a b

d c

B

1 2 3 4 5 6

CONTROL CYCLODEXTRIN

1 2 3 4 5 6

cav

TrkA

Ras

Shc

figure 4

Top bottom Top bottom


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

C Xy CD F C Xy CD F

EGF

P-MAPK

MAPK

A B

C

C Xy CD F C Xy CD F F+m CD+col

NGF

C Xy CD F C Xy CD F

P-TrkA

TrkA

P-MAPK

MAPK

NGF

figure 5


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

PC12 cells have caveolae that contain TrkA. Caveolae-disrupting drugs inhibit NGF-, but not EGF-, induced MAPK phosphorylation

Documents