International Journal of

Molecular Sciences

Article
Nerve Growth Factor Signaling from Membrane
Microdomains to the Nucleus: Differential
Regulation by Caveolins
Ambre Spencer 1,2,3,† , Lingli Yu 1,2,3,† , Vincent Guili 2,† , Florie Reynaud 4 , Yindi Ding 1,2,3 ,
Ji Ma 1,3 , Jérôme Jullien 2 , David Koubi 2 , Emmanuel Gauthier 2 , David Cluet 1,2 , Julien Falk 4 ,
Valérie Castellani 4 , Chonggang Yuan 1,3, * and Brian B. Rudkin 1,2,5, *
1 East China Normal University, Key Laboratory of Brain Functional Genomics of the Ministry of Education of
PR China, Joint Laboratory of Neuropathogenesis, ECNU, ENS Lyon, CNRS, Shanghai 200062, China;
ambre.spencer@gmail.com (A.S.); liliyunet@gmail.com (L.Y.); emeline.d@gmail.com (Y.D.);
jma@bio.ecnu.edu.cn (J.M.); david.cluet@ens-lyon.fr (D.C.)
2 Univ. Lyon, Ecole normale supérieure de Lyon, Université Claude Bernard Lyon 1, CNRS,
Differentiation & Cell Cycle Group, Laboratoire de Biologie Moléculaire de la Cellule, UMR5239,
69007 Lyon, France; vincent.guili@ac-lyon.fr (V.G.); jj256@gurdon.cam.ac.uk (J.J.);
david.koubi@finovatis.com (D.K.); Emmanuel.GAUTHIER@ch-le-vinatier.fr (E.G.)
3 East China Normal University, School of Life Sciences, Laboratory of Molecular
and Cellular Neurophysiology, Shanghai 200062, China
4 Univ. Lyon, Université Claude Bernard Lyon 1, CNRS, CGphiMC UMR5534, 69622 Villeurbanne Cedex,
France; florie.reynaud@univ-lyon1.fr (F.R.); julien.falk@univ-lyon1.fr (J.F.);
valerie.castellani@univ-lyon1.fr (V.C.)
5 Univ. Lyon, Université Claude Bernard Lyon 1, Inserm, Stem Cell and Brain Research Institute U1208,
69500 Bron, France
* Correspondences: cgyuan@bio.ecnu.edu.cn (C.Y.); brian.rudkin@inserm.fr (B.B.R.);
Tel.: +86-216-223-2729 (C.Y.); +33-472-913-475 (B.B.R.); Fax: +86-216-223-2729 (C.Y.); +33-472-913-461 (B.B.R.)
† These authors contributed equally to this work.

Academic Editors: Margaret Fahnestock and Keri Martinowich

Received: 29 December 2016; Accepted: 13 March 2017; Published: 24 March 2017

Abstract: Membrane microdomains or “lipid rafts” have emerged as essential functional modules
of the cell, critical for the regulation of growth factor receptor-mediated responses. Herein we
describe the dichotomy between caveolin-1 and caveolin-2, structural and regulatory components of
microdomains, in modulating proliferation and differentiation. Caveolin-2 potentiates while caveolin-1
inhibits nerve growth factor (NGF) signaling and subsequent cell differentiation. Caveolin-2 does not
appear to impair NGF receptor trafficking but elicits prolonged and stronger activation of MAPK
(mitogen-activated protein kinase), Rsk2 (ribosomal protein S6 kinase 2), and CREB (cAMP response
element binding protein). In contrast, caveolin-1 does not alter initiation of the NGF signaling
pathway activation; rather, it acts, at least in part, by sequestering the cognate receptors, TrkA and
p75NTR , at the plasma membrane, together with the phosphorylated form of the downstream effector
Rsk2, which ultimately prevents CREB phosphorylation. The non-phosphorylatable caveolin-1
serine 80 mutant (S80V), no longer inhibits TrkA trafficking or subsequent CREB phosphorylation.
MC192, a monoclonal antibody towards p75NTR that does not block NGF binding, prevents exit of
both NGF receptors (TrkA and p75NTR ) from lipid rafts. The results presented herein underline the
role of caveolin and receptor signaling complex interplay in the context of neuronal development
and tumorigenesis.

Keywords: growth factor signaling; NGF; Trk; p75NTR ; trafficking; lipid rafts; membrane microdomains;
caveolin; PC12; CREB; dorsal root ganglion neurons

Int. J. Mol. Sci. 2017, 18, 693; doi:10.3390/ijms18040693 www.mdpi.com/journal/ijms

Int. J. Mol. Sci. 2017, 18, 693 2 of 25

1. Introduction
Membrane microdomains, rich in cholesterol, sphingomyelins, and glycolipids, called lipid rafts,
are functional modules of the cell membrane that play a key role in regulating cellular responses to
environmental stimuli—e.g., the presence or absence of growth factors or infectious agents—through
their capacity to attract or deploy select cellular components [1]. Some domains are characterized
by the presence of specific structural and regulatory proteins called caveolins, which, when present
in appropriate amounts and with specific post-translational modification, can form caveolae [2,3].
The caveolin family consists of two ubiquitously expressed genes, caveolin-1 (Cav-1) and -2 (Cav-2),
and one specifically expressed in smooth and skeletal muscles, caveolin-3 (Cav-3) [4,5]. While
expression of Cav-1 or Cav-3 is sufficient for the formation of caveolae, expression of Cav-2 is not [6–8].
Cav-1 interacts with numerous proteins involved in signal transduction, which could result in
compartmentalization of signaling molecules and their maintenance in an inactive conformation [2,9].
Caveolins influence many diverse physiological processes through their localization in lipid rafts, their
role in formation of caveolae, and their specific interaction with both lipid and protein components
thereof [10].
At the cell membrane, nerve growth factor receptors TrkA and p75NTR are concentrated in different
subtypes of lipid rafts including caveolae [11–13]. In addition to its localization in caveolae, the nerve
growth factor (NGF)-receptor TrkA was shown to interact with Cav-1 [14]. Consistently, perturbation
of caveolae and modification of Cav-1 expression, alters NGF signaling in different cell types [11,14,15].
How caveolins impact NGF signaling from TrkA remains to be clarified. This is particularly interesting
considering the different role of Cav-1 and Cav-2 in caveolae formation and their differential regulation
during NGF-induced PC12 differentiation [16].
Initially difficult to detect, e.g., [17], caveolins have been observed in brain and several
neuronal cell types, playing key roles in neuronal signaling, underlined by correlative and functional
observations that certain neurodegenerative diseases may have links to microdomain components
including Cav-1 [18–20]. Similarly, mutations, loss of, or overexpression of Cav-1 have been associated
with numerous cancers [21–24], resulting in it being considered a bio-marker and potential therapeutic
target [25–27]. Indeed, Cav-1 is emerging as a key component of circulating exosomes and is elevated,
e.g., in melanoma [28] and glioblastoma [29], while its expression in target cells plays a regulatory role
in exosome internalization, essential for transmitting molecular information participating in tumor
development [30,31].
The present article offers more insight into these issues. Cav-1 overexpression inhibits neurite
growth from neurons of the dorsal root ganglia. Cav-1 and Cav-2 differentially modulate NGF signaling
events in the PC12 model. Notably, Cav-2 potentiates NGF signaling and the resulting physiological
response. By contrast, Cav-1 inhibits NGF signaling without, however, impairing transient MAPK
(mitogen-activated protein kinase) pathway activation. Rather, it acts by sequestering NGF receptors
and downstream effector, phosphorylated-Rsk2, at the plasma membrane, resulting in the prevention
of the phosphorylation of CREB (cAMP response element binding protein). Some insight into the
mechanism of action of Cav-1 is afforded by the study of the non-phosphorylatable Cav-1 mutant
(S80V), which no longer inhibits TrkA trafficking or CREB phosphorylation. In summary, our results
contribute to an understanding of the impact of TrkA and caveolin interplay in tumorigenesis and
neuronal differentiation.

2. Results

2.1. Caveolin-1 (Cav-1) Overexpression Impairs Neurite Growth of Embryonic Sensory Neurons
Sensory neurons from dorsal root ganglia (DRG) express a combination of Trk neurotrophin
receptors during development, whereas in adult, the different receptors are mainly expressed
individually in discrete sub-populations [32]. NGF was shown to regulate morphology and growth
of DRG axons in vitro [33–35]. Expression studies were performed in embryonic DRG neurons from
Int. J. Mol. Sci. 2017, 18, 693 3 of 25

stage E14.5 cultured on laminin substratum. At this stage, TrkA is expressed in nearly 80% of the DRG
neurons and NGF was shown to enhance laminin-induced outgrowth [35,36]. Consistent with previous
reports in rat neurons, we found that Cav-1 is expressed by TrkA-positive DRG neurons and is found
both in the soma and the neurites [16] (Figure 1A). To get efficient transfection with low amounts of
Int. J. Mol. Sci. 2017, 18, 693 3 of 24
cells, electroporation of DRG neurons was optimized using the Neon® transfection system [37]. Using
this procedure,
previous wereports
transfected more than
in rat neurons, 50%that
we found of the neurons
Cav-1 (Figure
is expressed 1B,C). We found
by TrkA-positive that many DRG
DRG neurons
and is foundwith
neurons transfected both in the soma and(red
Cav-1-RFP the neurites [16] (Figure
fluorescent 1A). [38]
protein) To getand
efficient
GFPtransfection
(green with low protein)
fluorescent
amounts of cells, electroporation of DRG neurons was optimized using the Neon® transfection system
differentiate (Figure 1D–G. Quantification of the total neurite length of transfected neurons showed
[37]. Using this procedure, we transfected more than 50% of the neurons (Figure 1B,C). We found
that neuronsthatco-expressing
many DRG neurons Cav-1-RFP
transfectedand GFP
with grew shorter
Cav-1-RFP neurites
(red fluorescent than RFP-
protein)[38] andand
GFPGFP-expressing
(green
ones. Indeed, average
fluorescent lengthdifferentiate
protein) decreased (Figure
by 18%1D–G.
in Cav-1 overexpressing
Quantification neurons
of the total after
neurite one of
length day in vitro
transfected neurons showed that neurons co-expressing Cav-1-RFP
(12 montages representative of 339 RFP-positive neurons and 293 Cav-1-RFP-expressing neurons, and GFP grew shorter neurites
than RFP- and GFP-expressing ones. Indeed, average length decreased by 18% in Cav-1
p = 0.018).overexpressing
In parallel with a progressive accumulation of Cav-1, this decrease reached 33% after two
neurons after one day in vitro (12 montages representative of 339 RFP-positive
days (434 RFP-positive and
neurons and 293 Cav-1-RFP-expressing neurons,p p= =0.018).
308 RFP-positive neurons, 0.00009) (Figurewith
In parallel 1H).a This effect was even
progressive
accumulation
stronger when the amount of Cav-1, this decrease
of plasmid reached 33%
is doubled, after two
further days (434aRFP-positive
supporting and 308effect.
dose-dependent RFP- Average
positive neurons, p = 0.00009) (Figure 1H). This effect was even stronger when the amount of plasmid
total neurite length dropped by 34% (361 RFP-positive and 416 RFP-positive neurons, p = 0.002) after
is doubled, further supporting a dose-dependent effect. Average total neurite length dropped by 34%
the first day and 41% on the second day (296 RFP-positive and 351 RFP-positive neurons, p = 0.000002).
(361 RFP-positive and 416 RFP-positive neurons, p = 0.002) after the first day and 41% on the second
Thus, overexpression
day (296 RFP-positive of Cav-1 andappears to impair
351 RFP-positive neurite
neurons, growth of
p = 0.000002). DRG
Thus, neurons as
overexpression of it does in PC12
Cav-1
appears
cells [14,16]. The to PC12impair neurite
line wasgrowth
used inof DRG neurons as it does
the subsequent in PC12 cells [14,16]. The PC12 line was
studies.
used in the subsequent studies.

Figure 1. Caveolin-1 (Cav-1) expression inhibits neurite outgrowth from mouse Dorsal root ganglia
Figure 1. Caveolin-1 (Cav-1) expression inhibits neurite outgrowth from mouse Dorsal root ganglia
(DRG) neurons in culture. Cav-1 is detected in both the soma and the neuritic processes of E14.5 DRG
(DRG) neurons
neuronsin culture.
(A); Cav-1 is leads
Neon® transfection detected in both
to efficient the somaofand
electroporation E14.5the
DRG neuritic
neuronsprocesses
with little of E14.5
DRG neurons (A);effects ®
Neon(B,C);transfection leads to efficient electroporation of E14.5
adverse Neurons co-expressing GFP (Green Fluorescent Protein) andDRG neurons with
RFP (Red
Fluorescent
little adverse effectsProtein)
(B,C);(D-D’’)
Neuronsor GFP and Cav-1-RFP (E–E”)
co-expressing GFP can differentiate
(Green in vitro. Phase
Fluorescent images
Protein) and(D RFP (Red
and E) exemplify the morphology of GFP (D’and E’) and RFP (D”) or Cav1-RFP (E”) expressing
Fluorescent Protein) (D–D”) or GFP and Cav-1-RFP (E–E”) can differentiate in vitro. Phase images
neurons. Nevertheless, neurons expressing Cav-1-RFP grew shorter processes than neurons
(D and E) expressing
exemplify the morphology of GFP (D’ and E’) and RFP (D”) or Cav1-RFP (E”) expressing
RFP (F,G); The length of GFP positive neurites measured and divided by the number of
neurons. Nevertheless, neurons
transfected neurons expressing
(H); Results Cav-1-RFP
are pooled from three grew shorter
sets of processes
cultures, each culturethan neurons
included fourexpressing
RFP (F,G); The length of GFP positive neurites measured and divided by the number of transfected
mosaic fields containing >250 transfected cells. Mean ± SEM; (** p < 0.01; *** p < 0.00001). Statistical
analysis
neurons (H); was performed
Results are pooled using the two-tail
from paired
three sets of Student’s
cultures,t-test.
eachScale bars represent
culture included 100four
µm in B,
mosaic fields
C, F and G, and 10 µm in A, D” and E’’.
containing >250 transfected cells. Mean ± SEM; (** p < 0.01; *** p < 0.00001). Statistical analysis was
performed using the two-tail paired Student’s t-test. Scale bars represent 100 µm in B, C, F and G, and
10 µm in A, D” and E”.
Int. J. Mol. Sci. 2017, 18, 693 4 of 25

2.2. Effect of Cav-1 and Cav-2 on Nerve Growth Factor (NGF)-Induced Neurite Formation in PC12 Cells
Interestingly, Cav-1 and Cav-2 exhibit different kinetics of expression during NGF-induced
differentiation. In agreement with previously-published studies [16], Cav-1 expression is almost
undetectable in sparse populations and progressively increases in PC12 cells in response to NGF.
In contrast, Cav-2 expression increases to a maximum after two days of treatment with NGF, then
drops precipitously (Figure S1). Since, Cav-1 and Cav-2 have an expression peak during late and early
phases of neuritogenesis, respectively, we wondered whether they could have different effects on
NGF-induced PC12 cell morphological differentiation.
In order to evaluate the impact of caveolin expression on global NGF signaling, normal PC12
cells were stably-transfected and assayed for transgene expression. Clones expressing Cav-1 or Cav-2
were selected for subsequent studies with most experiments being performed on two or more clones
(Table S1). NGF induces PC12 cells to stop proliferating and to differentiate into a sympathetic
neuron-like phenotype [39]. Normal PC12 cells, Cav-1 PC12 cells and Cav-2 PC12 cells were plated
at low density and exposed for three days to NGF concentrations ranging from 5 to 50 ng/mL.
In response to NGF, normal PC12 cells extended a characteristic neurite network in a dose-dependent
manner whereas Cav-1 PC12 cells did not elicit any significant process extension at any of the NGF
concentrations evaluated (Figure 2A), a result consistent with previous studies [14,16]. In contrast,
Cav-2 PC12 cells exposed to NGF developed a more extensive and denser neurite network than
normal PC12 cells (Figure 2A). NGF-induced morphological differentiation of Cav-2 PC12 cells thus
appears to be potentiated in comparison to the response of normal PC12 cells. The global evaluation
of differentiation of each cell line at the different NGF concentrations is represented in Figure 2B.
These observations suggest that Cav-1 and Cav-2 have opposite effects on NGF-induced PC12 cell
morphological differentiation. Corroborating this observation, exposure of the PC12 cells to siRNA
towards Cav-1 resulted in an increase of differentiation while siRNA to Cav-2 would appear to
have had a minimal effect (Figure 2C). The extremely low levels of expression of Cav-1 and Cav-2
in exponentially growing PC12 cells made it difficult to appreciate the relative impact of the two
siRNA on the expression of their respective targets. These results are therefore to be taken with the
necessary caution.

2.3. Effect of Cav-1 and Cav-2 on the Anti-Mitogenic Response to NGF

PC12 cells respond to NGF by induction of an anti-mitogenic response [39] elicited by cell cycle
arrest in the G1 phase [40], and the triggering of differentiation. Normal PC12 cells, Cav-1 PC12 cells
and Cav-2 PC12 cells were plated at low density in the presence of serum and grown for two days
prior the addition of 20 ng/mL NGF. In the presence of NGF, normal PC12 cells and Cav-2 PC12 cells
undergo a final round of division and stop proliferating with similar kinetics (Figure 3A). It would
appear that the potentiation of NGF-induced differentiation, observed with Cav-2 overexpression,
is not a consequence of an enhanced anti-mitogenic response. In contrast, Cav-1 PC12 cells did not stop
proliferating after NGF addition and showed a growth rate similar to that of exponentially-growing
normal PC12 cells in absence of NGF (Figure 3A). The anti-mitogenic effect of NGF is propagated via
the induction of the cyclin-dependent kinase inhibitor p21WAF/Cip1 [41,42]. To test if Cav-1 or Cav-2
expression alters p21WAF/Cip1 levels, equal numbers of cells exposed to 20 ng/mL of NGF for 0, 1,
or 3 days were lysed and protein expression ascertained by western analysis. Figure 3B shows that
p21WAF/Cip1 levels in normal PC12 cells and in Cav-2 PC12 cells markedly increase during the time
course of NGF treatment. No change in p21WAF/Cip1 level was detectable in Cav-1 PC12 cells exposed
to NGF.
Int. J. Mol. Sci. 2017, 18, 693 5 of 25
Int. J. Mol. Sci. 2017, 18, 693 5 of 24

Figure 2. Effect of Cav-1 and Cav-2 expression on NGF-induced PC12 cell differentiation. (A) Normal
Figure 2. Effect of Cav-1 and Cav-2 expression on NGF-induced PC12 cell differentiation. (A) Normal
PC12 cells, stably-transfected Cav-1 PC12 cells Cav-2-PC12 or Cav-1 S80V PC12 cells were plated at
PC12 cells, stably-transfected Cav-1 PC12 cells Cav-2-PC12 or Cav-1 S80V PC12 cells were plated at a
a density of 105 cells/well on collagen/poly-lysine coated six-well plates. Cells were maintained in
density of 10 cells/well on collagen/poly-lysine coated six-well plates. Cells were maintained in
5
DMEM (Dulbecco’s Modified Eagle’s medium) supplemented with serum for 24 h After 17 h of serum
DMEM (Dulbecco’s Modified Eagle’s medium) supplemented with serum for 24 h After 17 h of serum
deprivation, NGF was added at increasing concentrations (0, 5, 20, 50 ng/mL NGF). After three days in
deprivation, NGF was added at increasing concentrations (0, 5, 20, 50 ng/mL NGF). After three days
culture, cells were photographed; (B) Analysis of neurite outgrowth. Quantification of the percentage
in culture, cells were photographed; (B) Analysis of neurite outgrowth. Quantification of the
of cells exhibiting neurites longer than three cell body diameters after 72 h of NGF treatment at 0, 5,
percentage of cells exhibiting neurites longer than three cell body diameters after 72 h of NGF
and 50 ng/mL. Approximately 100 cells per condition from five independent experiments were taken
treatment at 0, 5, and 50 ng/mL. Approximately 100 cells per condition from five independent
into account. Values are mean percentage ± SEM. Statistical significance of the observations between
experiments were taken into account. Values are mean percentage ± SEM. Statistical significance of
PC12 cells and Cav mutant (Cav-1; Cav-2; Cav-1 S80V) PC12 cells are indicated by * p < 0.05 (unpaired,
the observations between PC12 cells and Cav mutant (Cav-1; Cav-2; Cav-1 S80V) PC12 cells are
two-tail Student’s t-test). (C) Effect of siRNA towards Cav-1 and Cav-2 on neurite outgrowth. Normal
indicated by * p < 0.05 (unpaired, two-tail Student’s t-test). (C) Effect of siRNA towards Cav-1 and
PC12 cells, plated on collagen/poly-lysine coated 24-well plates (2 × 104 cells per well), were transiently
Cav-2 on neurite outgrowth. Normal PC12 cells, plated on collagen/poly-lysine coated 24-well plates
transfected with 30 nM siRNA against caveolin-1; caveolin-2 and FITC (fluoresceine isothiocyanate)
(2 × 104 cells per well), were transiently transfected with 30 nM siRNA against caveolin-1; caveolin-2
conjugated scrambled siRNA. Cells were maintained in DMEM with 0 or 20 ng/mL NGF. After 3
and FITC (fluoresceine isothiocyanate) conjugated scrambled siRNA. Cells were maintained in
days, cells were photographed and neurite outgrowth was quantified by ImageJ (Average 400 cells per
DMEM with 0 or 20 ng/mL NGF. After 3 days, cells were photographed and neurite outgrowth was
condition compiled from two independent experiments. Values are mean ± SEM. Statistical significance
quantified
of the by ImageJ
observations (Average
between 400 cells
PC12 cells per condition compiled
and Cav-expressing from two
(Cav-1; Cav-2) PC12 independent experiments.
cells are indicated by
Values are mean ± SEM. Statistical significance of the observations between PC12
+ p < 0.005; ++ p < 0.0005; +++ p < 0.00005 (% cells with neurites) and # p < 0.0005; ### p < 0.0000001cells and Cav-
expressing
(Average (Cav-1;
neurite Cav-2)
length) PC12 cellsby
as ascertained arethe
indicated by two-tail
unpaired, + p < 0.005; ++ p < 0.0005;
Student’s t-test). +++ p < 0.00005 (% cells
with neurites) and # p < 0.0005; ### p < 0.0000001 (Average neurite length) as ascertained by the
unpaired, two-tail Student’s t-test).
Int. J. Mol. Sci. 2017, 18, 693 6 of 25
Int. J. Mol. Sci. 2017, 18, 693 6 of 24

Figure3.3.Effect
Figure EffectofofCav-1
Cav-1andandCav-2
Cav-2expression
expressionon onthe
theanti-mitogenic
anti-mitogeniceffect
effectofofNGF.
NGF.(A) (A)Normal
NormalPC12 PC12
cells, 2
cells,Cav-1
Cav-1PC12PC12cells,
cells,Cav-2
Cav-2PC12,
PC12,andandCav-1
Cav-1S80V
S80VPC12PC12cells
cellswere
wereplated
platedatata alow
lowdensity
densityinin2525cmcm2
dishes
dishescontaining
containingaamedium
mediumsupplemented
supplemented with with serum. NGF 20
serum. NGF 20 ng/mL
ng/mLwas wasadded
addedtotothe themedium
medium on
on day 2, except for one series of dishes that was left untreated. Cell number was
day 2, except for one series of dishes that was left untreated. Cell number was counted every day for counted every day
for
sixsix days.The
days. Theexperiment
experimentwas wasperformed
performedthree
threetimes,
times, with
with two
two different
different clones
clonesof ofCav-1
Cav-1PC12
PC12and and
Cav-2-PC12 cells. Cell number is normalized to cell number at day 1 (N1). Values are mean±±SEM;
Cav-2-PC12 cells. Cell number is normalized to cell number at day 1 (N1). Values are mean SEM;
(B)
(B)Normal
NormalPC12 PC12cells,
cells,Cav-1
Cav-1PC12
PC12cells,
cells,and
andCav-2
Cav-2 PC12
PC12 cells
cells were
were exposed
exposed to to 20 ng/mLNGF
20 ng/mL NGFfor for0,
0,1,1,or
or 33 days.
days. Proteins
Proteins were
were extracted and resolved by SDS-PAGE and immunoblotted
extracted and resolved by SDS-PAGE and immunoblotted with CP36 with CP36
monoclonal WAF1/Cip1 Equal loading was verified by reprobing the same
monoclonalantibody
antibody directed against
directed againstp21p21WAF1/Cip1.. Equal loading was verified by reprobing the same
blot WAF1/Cip1
blotwith
withanti-Histone
anti-Histone H1H1 antibody;
antibody; (C)(C)
The NGF-dependent
The NGF-dependent induction
inductionof p21
of p21WAF1/Cip1was wasanalyzed
analyzed
ininnormal PC12 cells, Cav-1 PC12 cells, and Cav-2 PC12 cells. Cells were transiently
normal PC12 cells, Cav-1 PC12 cells, and Cav-2 PC12 cells. Cells were transiently transfected transfected with with
the
minimal
the minimalp21 promoter–Luciferase
p21 promoter–Luciferase reporter (p21–Luc)
reporter and treated
(p21–Luc) or not with
and treated or notNGFwith(50 ng/mL
NGF for 48 h)
(50 ng/mL for
as described in the Materials and methods section (Mean ± SEM of three independent experiments.
48 h) as described in the Materials and methods section (Mean ± SEM of three independent
Statistical significance was determined using a regular two-way ANOVA test with Bonferroni post hoc
experiments. Statistical significance was determined using a regular two-way ANOVA test with
tests ** p < 0.01 versus the corresponding–NGF group. ## p < 0.01, ### p < 0.001 versus the Cav-1 PC12
Bonferroni post hoc tests ** p < 0.01 versus the corresponding–NGF group. ## p < 0.01, ### p < 0.001
+ NGF group).
versus the Cav-1 PC12 + NGF group).
Int. J. Mol. Sci. 2017, 18, 693 7 of 25

The minimal p21WAF/Cip1 –promoter luciferase construct (p2193S-Luc) was used as reporter gene
for the NGF signaling pathway [43–45]. In normal PC12 cells treated with NGF for 48 h, the promoter is
activated asInt.ascertained by an increase in firefly luciferase activity. This activation of the p21
J. Mol. Sci. 2017, 18, 693 7 of 24
promoter
is also found in Cav-2 PC12 cells. In contrast, NGF-induced p21 promoter activation is reduced in
Cav-1 PC12 cells The(Figure
minimal3C). p21WAF/Cip1–promoter luciferase construct (p2193S-Luc) was used as reporter gene
for the NGF signaling pathway [43–45]. In normal PC12 cells treated with NGF for 48 h, the promoter
These results indicate that Cav-1, but not Cav-2 expression results in inhibition of the
is activated as ascertained by an increase in firefly luciferase activity. This activation of the p21
anti-mitogenic
promotereffectis of NGF,
also foundatinleast
Cav-2inPC12
part,cells.
by impairing activation ofp21
In contrast, NGF-induced transcription of p21isWAF/Cip1 .
promoter activation
reduced in Cav-1 PC12 cells (Figure 3C).
2.4. Effect of Cav-1 and
These Cav-2
results on NGF-Induced
indicate TrkA
that Cav-1, but notand NTR Internalization
p75expression
Cav-2 results in inhibition of the anti-
mitogenic effect of NGF, at least in part, by impairing activation of transcription of p21WAF/Cip1.
NGF receptor trafficking is essential for regulating many of the subsequent cellular
2.4. Effect of Cav-1
responses [13,46–55]. Theand Cav-2ofonCav-1
effect NGF-Induced TrkA and
and Cav-2 p75NTR Internalization
expression on TrkA was monitored in clones of
PC12 cells stably NGF expressing these proteins.
receptor trafficking Following
is essential for regulatingNGF
many treatment, Figure
of the subsequent 4A responses
cellular shows that TrkA
and p75NTR[13,46–55].
exit from The effectrafts
lipid of Cav-1 and Cav-2
in normal expression
PC12 on TrkAPC12
and Cav-2 was monitored
cells. In in clones of TrkA
contrast, and p75NTR
PC12 cells
stably expressing these proteins. Following NGF treatment, Figure 4A shows that TrkA and p75NTR
remain in lipid rafts in Cav-1 expressing cells, indicating that Cav-1 is retaining NGF receptors in lipid
exit from lipid rafts in normal PC12 and Cav-2 PC12 cells. In contrast, TrkA and p75NTR remain in
rafts. Quantification
lipid rafts inof several
Cav-1 independent
expressing experiments
cells, indicating that Cav-1(Figure 4B)NGF
is retaining shows that Cav-1
receptors in lipidalmost
rafts. totally
inhibits the exit of TrkA and p75
Quantification of several NTR from lipid rafts, whereas Cav-2 does not.
independent experiments (Figure 4B) shows that Cav-1 almost totally
inhibits the exit of TrkA and p75NTR from lipid rafts, whereas Cav-2 does not.

Figure 4. Effect of Cav-1 and Cav-2 expression on NGF receptor exit from lipid rafts. (A) TrkA and
Figure 4. Effect of Cav-1 and Cav-2 expression on NGF receptor exit from lipid rafts. (A) TrkA and
p75NTR levels in the lipid raft fraction (LRF) before and after addition of NGF (20 ng/mL for 45 min)
p75NTR levels in the lipid
to cultures raftPC12,
of normal fraction
Cav-1(LRF)
PC12,before and after
and Cav-2-PC12 addition
cells isolated,of as NGF (20 inng/mL
described for 45 min) to
the Materials
cultures of normal PC12, Cav-1 PC12, and Cav-2-PC12 cells isolated, as described in the Materials and
and Methods. Lipid raft fractions were then subjected to Western analysis. Nitrocellulose membranes
were probed
Methods. Lipid with RTA,
raft fractions wereanti-p75 NTR and anti-flotillin-1 antibodies. Flotillin-1 was used as a loading
then subjected to Western analysis. Nitrocellulose membranes were
control. The Odyssey imaging system was used for quantitative infrared fluorescence detection of the
probed with RTA, anti-p75NTR and anti-flotillin-1 antibodies. Flotillin-1 was used as a loading control.
relative amount of proteins; (B) Analysis of TrkA and p75NTR exit from lipid rafts. Level of TrkA and
The Odyssey p75imaging
NTR system was used for quantitative infrared
in lipid rafts was normalized for flotillin-1 level for each sample.fluorescence detection
Systematic of the
comparison of relative
amount of proteins;
data with and(B) without
Analysis of TrkA
flotillin correction gaveNTR
and p75 exit from
identical results.lipid
TrkArafts.
and p75Level of from
NTR exit TrkAtheand p75NTR in
lipid
raft fraction was then analyzed and represented as percent of the TrkA in lipid
lipid rafts was normalized for flotillin-1 level for each sample. Systematic comparison of data with and rafts compared to the
amount observed in the absence of NGF, considered as 100 percent. (Mean ± SEM of three
without flotillin correction gave identical results. TrkA and p75NTR exit from the lipid raft fraction was
independent experiments. Statistical significance of the effect of addition of NGF vs. the absence of
then analyzedNGF, and
wasrepresented as percent
ascertained using of the
the unpaired TrkAt-Test.
Student’s in lipid
* p <rafts compared
0.05, ** p < 0.01.) to the amount observed
in the absence of NGF, considered as 100 percent. (Mean ± SEM of three independent experiments.
Statistical significance of the effect of addition of NGF vs. the absence of NGF, was ascertained using
the unpaired Student’s t-Test. * p < 0.05, ** p < 0.01.)
Int. J. Mol. Sci. 2017, 18, 693 8 of 25

Ligand-induced internalization of TrkA and subcellular localization of TrkA were evaluated

using confocal microscopy (Figure 5). Normal PC12 cells, Cav-1 PC12 cells and Cav-2 PC12 cells
Int. J. Mol. Sci.
were transiently 2017, 18, 693 with a chimeric TrkA-EGFP receptor [55]. Expression of8 this
transfected of 24 receptor,

together with the use of RTAinternalization

Ligand-induced antibody as an NGFand
of TrkA agonist [55,56]
subcellular allow simultaneous
localization of TrkA were monitoring
evaluated of two
different pools of TrkA in a single cell. (i) “TrkA-EGFP” reflects EGFP fluorescence,
using confocal microscopy (Figure 5). Normal PC12 cells, Cav-1 PC12 cells and Cav-2 PC12 cells were which corresponds
transiently
to all of the cellulartransfected
localizations with a chimeric TrkA-EGFP(ii)
of TrkA-EGFP; receptor
“Cell [55]. Expression
surface TrkA of this
at receptor,
t = 0 min” together
is identified
with the use of RTA ◦ antibody as an NGF agonist [55,56] allow simultaneous monitoring of two
by incubation of cells at 4 C for 30 min in presence of RTA, followed by fixation, permeabilization
different pools of TrkA in a single cell. (i) “TrkA-EGFP” reflects EGFP fluorescence, which
and labeling of cells to
corresponds with
all ofrhodamine-conjugated anti-rabbit(ii)antibody.
the cellular localizations of TrkA-EGFP; “Cell surface This
TrkA allows
at t = 0 the
min”detection
is of
TrkA-RTA identified
complexes, which areofinitially ◦ C, it is possible
by incubation cells at 4at°C the
forcell
30 surface. Following
min in presence of RTA,thefollowed
shift to by37 fixation,
to monitorpermeabilization
the fate of theand labeling ofthat
receptors cells have
with rhodamine-conjugated
moved from the cell anti-rabbit
surface antibody. This allows locations
to intracellular
the detection of TrkA-RTA complexes, which are initially at the cell surface. Following the shift to 37
during the incubation time. As shown in Figure 5, first row, incubation of PC12 cells in the presence of
°C, it is possible to monitor the fate of the receptors that have moved from the cell surface to
RTA at 4 ◦ C does notlocations
intracellular allow receptor
during theinternalization,
incubation time. Asreflected by the5,presence
shown in Figure of RTA-TrkA
first row, incubation of PC12complexes
exclusivelycells
at inthe cell surface. Incubation of PC12 cells and of Cav-2 PC12
the presence of RTA at 4 °C does not allow receptor internalization, reflected by the presencecells for 20 min at 37 ◦ C
of RTA-TrkA complexes exclusively at the cell surface. Incubation
in the presence of RTA leads to partial redistribution of the TrkA–antibody complex inside of PC12 cells and of Cav-2 PC12 the cells.
cells for 20 min at 37 °C in the presence of RTA leads to partial redistribution of the TrkA–antibody
Intracellular RTA–TrkA complexes correspond to internalized receptors. By contrast, following the
complex
◦ C of Cav-1 inside the cells. Intracellular RTA–TrkA complexes correspond to internalized receptors. By
shift to 37 contrast, PC12
following thecells treated
shift to 37 °C of with
Cav-1 RTA,
PC12little or no intracellular
cells treated with RTA, littleRTA-TrkA complexes are
or no intracellular
detected, indicating
RTA-TrkA complexes that TrkAare internalization
detected, indicatingis impaired
that TrkA in internalization
Cav-1 overexpressingis impairedcells.
in Cav-1
overexpressing cells.

Figure 5. Effect of Cav-1 and Cav-2 expression on TrkA trafficking. Internalization of TrkA was
Effect ofbyCav-1
Figure 5. provoked and Cav-2 expression on TrkA trafficking. Internalization of TrkA was
RTA addition to normal PC12, Cav-1 PC12, Cav-2 PC12, and Cav-1 S80V PC12 cells
provoked transiently
by RTA addition to normalchimerae
expressing TrkA–EGFP PC12, Cav-1 PC12,
(green), as Cav-2
described PC12,
in the and
Materials andCav-1 S80V
Methods. AfterPC12 cells
transientlyfixation
expressing TrkA–EGFP chimerae
and permeabilization of the cell,(green), as described
RTA localization was in the Materials
determined using aand Methods. After
secondary
antibody
fixation and labeled with rhodamine
permeabilization (red)
of the cell, RTAat the indicated times.
localization wasRhodamine
determined fluorescence
using a indicates
secondarythe antibody
location of TrkA–RTA and TrkA–EGFP–RTA complexes that were at the cell surface at the beginning
labeled with rhodamine (red) at the indicated times. Rhodamine fluorescence indicates the location
of the experiment (Cell Surface TrkA at t = 0 min). Distribution of cell surface TrkA (at t = 0) in Cav-1
of TrkA–RTAPC12,and TrkA–EGFP–RTA
Cav-2-PC12, and Cav-1 S80Vcomplexes that
PC12 cells after were
0 min at°Cthe
at 37 cell surface
is identical at the beginning
to that observed in PC12 of the
(Cell Surface TrkA at t = 0 min). Distribution of cell surface TrkA (at t = 0) in Cav-1 PC12,
experimentcells.
Cav-2-PC12, and Cav-1 S80V PC12 cells after 0 min at 37 ◦ C is identical to that observed in PC12 cells.

Thus, the differential internalization of TrkA in the presence of increased levels of Cav-1 and Cav-2
appears to mirror the functional outcome of Cav-1 and Cav-2 overexpression on PC12 differentiation.
Int. J. Mol. Sci. 2017, 18, 693 9 of 24

Thus, the differential internalization of TrkA in the presence of increased levels of Cav-1 and
Int. J. Mol. Sci. 2017, 18, 693 9 of 25
Cav-2 appears to mirror the functional outcome of Cav-1 and Cav-2 overexpression on PC12
differentiation.
2.5. Interplay of p75NTR and TrkA
2.5. Interplay of p75NTR and TrkA
When PC12 cells are treated with MC192 monoclonal antibody, specific for the extracellular
domainWhen of PC12
p75NTRcells[57],
are treated with MC192 monoclonal
TrkA internalization antibody,
is inhibited, specific
even in for the
presence of extracellular
high NGF
domain of p75 NTR [57], TrkA internalization is inhibited, even in presence of high NGF concentrations
concentrations [58]. Figure 6A (characteristic Western Blot) and Figure 6B (quantitation of multiple
[58]. Figure 6A
experiments) (characteristic
illustrate Western
that exposure of Blot)
PC12andcellsFigure
to this6B (quantitation
anti-p75 of multiple
NTR antibody almostexperiments)
completely
illustrate that
abolishes exposure ofTrkA
NGF-induced PC12and
cells
p75to thisexit
NTR anti-p75 NTR
from lipidantibody almostlike
rafts, much completely abolishes NGF-
Cav-1 expression. This
induced TrkA and p75 NTR exit from NTR
suggests that part of the role of p75 lipid rafts, much like Cav-1 expression. This suggests that part
in TrkA internalization may be to participate in, or regulate, of
the role of p75 NTR in TrkA internalization may be to participate in, or regulate, the exit from lipid rafts
the exit from lipid rafts in response to NGF.
in response to NGF.

Figure 6.
Figure 6. Effect
EffectofofMC192
MC192on onTrkA
TrkA exit
exit from
from lipid
lipid rafts
rafts before
before andandafterafter addition
addition of NGF.
of NGF. (A) Lipid
(A) Lipid rafts
rafts isolated
were were isolated from normal
from normal PC12treated
PC12 cells cells treated
with NGF with(20NGF
ng/mL(20 ng/mL for 45
for 45 min), min), with
treated treated NTR
p75with
p75NTR MC192
MC192 monoclonalmonoclonal
antibodyantibody
(8 ng/mL) (8 for
ng/mL)
30 min forprior
30 min prior
to and to and
during NGFduring NGF(20
exposure exposure
ng/mL for(20
ng/mL
45 min) for
and45from
min) and fromnormal
untreated untreated
PC12normal PC12
cells were cells were
isolated isolated in
as described asthe
described
Materials in and
the Materials
Methods.
and Methods.
Lipid rafts wereLipid
then rafts weretothen
subjected subjected
Western to Western
analysis. analysis.
Nitrocellulose Nitrocellulose
membranes were membranes
probed with were
RTA
probed
and with RTA and
anti-flotillin-1 anti-flotillin-1
antibody to detectantibody to detect and
TrkA expression TrkAflotillin-1
expression was and flotillin-1
used was used
as a loading as a
control.
Revelation was achieved
loading control. Revelationusing
wasquantitative
achieved usinginfrared fluorescence
quantitative detection
infrared using thedetection
fluorescence Odysseyusing
imagingthe
system; (B) Analysis of TrkA exit from lipid rafts. Level of TrkA in the lipid rafts
Odyssey imaging system; (B) Analysis of TrkA exit from lipid rafts. Level of TrkA in the lipid rafts was normalized to the
Flotillin-1 level for
was normalized to each sample. TrkA
the Flotillin-1 level exit from sample.
for each the lipidTrkA
raft fraction
exit from (LRF) was then
the lipid analyzed(LRF)
raft fraction and
represented as percent
was then analyzed and of the TrkA in
represented as lipid
percentrafts vs. TrkA in lipid
of the normal raftsPC12 cells considered
vs. TrkA in normal PC12as 100%.
cells
(Mean ± SEM
considered as of four (Mean
100%. independent
± SEMexperiments. Statisticalexperiments.
of four independent significance Statistical
vs. cells insignificance
the absence vs.
of NGF,
cells
was ascertained using the unpaired, two-tail Student’s t-Test. ** p < 0.01.)
in the absence of NGF, was ascertained using the unpaired, two-tail Student’s t-Test. ** p < 0.01.)

2.6. Effect of Cav-1 and Cav-2 on Activation of

2.6. of TrkA
TrkAand
andDownstream
DownstreamSignaling
SignalingEffectors
Effectors

NGF signaling
signalingandandPC12
PC12 cellcell
differentiation are mediated
differentiation are mediated via TrkA activation
via TrkA [59]. Since
activation [59].receptor
Since
endocytosis
receptor is important
endocytosis to orchestrate
is important NGFNGF
to orchestrate signaling,
signaling, we we
tested
testedwhether
whetherCav-1
Cav-1 and Cav-2
and Cav-2
differentially regulate
differentially regulate the
the canonical
canonicalNGFNGFsignaling
signalingcascade
cascadevia viaRas/Raf/MAPK.
Ras/Raf/MAPK. Normal
Normal PC12 cells,
Cav-1 PC12
Cav-1 PC12 cells
cells and
and Cav-2
Cav-2 PC12
PC12 cells
cells were
wereexposed
exposedto to2020ng/mL
ng/mL NGF for 10 min. In the absence of of
NGF, no
NGF, no TrkA
TrkAphosphorylation
phosphorylation was was detectable
detectable in
in any
any cell
cell line.
line. However, 10 10 min
min exposure
exposure to to NGF
NGF
resulted in
resulted in marked
marked tyrosine
tyrosine phosphorylation
phosphorylation in all all cell
cell lines,
lines, with
with an
an increased
increased phosphorylation
phosphorylation of of
TrkA in Cav-1 and Cav-2 PC12 cells compared to TrkA phosphorylation in normal PC12 cells (Figure 7A
(characteristic western blot) and Figure 7B (quantitation of multiple experiments).
Int. J. Mol. Sci. 2017, 18, 693 10 of 24

Int. J. Mol. Sci. 2017, 18, 693 10 of 25

TrkA in Cav-1 and Cav-2 PC12 cells compared to TrkA phosphorylation in normal PC12 cells (Figure
7A (characteristic western blot) and Figure 7B (quantitation of multiple experiments).

7. Effect
Figure 7. Effectof
ofCav-1
Cav-1andandCav-2
Cav-2 expression
expression ononTrkA activation.
TrkA Normal
activation. PC12
Normal cells,cells,
PC12 Cav-1 PC12PC12
Cav-1 cells
and
cellsCav-2 PC12 PC12
and Cav-2 cells were
cells left
wereuntreated or treated
left untreated with NGF
or treated with20NGF
ng/mL for 10 min
20 ng/mL after
for 10 min17after
h of 17
serum
h of
deprivation. (A) One mg protein of cell lysate was immunoprecipitated with
serum deprivation. (A) One mg protein of cell lysate was immunoprecipitated with a polyclonal a polyclonal antibody
directed
antibodyagainst TrkA
directed C-terminal
against domain (anti-Trk
TrkA C-terminal domain C-14), and immunoprecipitates
(anti-Trk were subjected
C-14), and immunoprecipitates wereto
analysis
subjectedbyto Western
analysisBlotby(WB). Membrane
Western Blot was
(WB).probed with an was
Membrane antibody to phosphotyrosine
probed with an antibody (α-PY)to
followed by a secondary
phosphotyrosine (α-PY)antibody
followedconjugated to horseradish
by a secondary peroxidase that
antibody conjugated allows chemiluminescence
to horseradish peroxidase that
detection; The same blot wasdetection;
allows chemiluminescence re-probedThe withsame
RTAblot
polyclonal antibodywith
was re-probed to TrkA
RTAfollowed
polyclonalby aantibody
secondary to
antibody
TrkA followed by a secondary antibody conjugated to a fluorochrome (B), which allowsusing
conjugated to a fluorochrome (B), which allows infrared fluorescence detection the
infrared
Odyssey imaging
fluorescence system
detection without
using the cross-reacting
Odyssey imaging withsystem
the firstwithout
antibodies (Mean ± SEM
cross-reacting withfrom
thethree
first
independent
antibodies (Meanexperiments.
± SEM from Statistical significance vs.
three independent PC12 cells Statistical
experiments. was ascertained using vs.
significance thePC12
unpaired,
cells
two-tail t-Test.
Student’susing
was ascertained p < 0.01). two-tail Student’s t-Test. ** p < 0.01).
the**unpaired,

Thus, Cav-1
Thus, Cav-1expression
expression doesdoes not appear
not appear to phosphorylation
to inhibit inhibit phosphorylation of TrkA
of TrkA under under the
the experimental
experimental conditions used herein. To further characterize the impact of Cav-1
conditions used herein. To further characterize the impact of Cav-1 and Cav-2, downstream effectors and Cav-2,
downstream
of effectorspathway
the NGF signaling of the NGFweresignaling pathway
monitored. were monitored.
Phosphorylated Phosphorylated
TrkA leads TrkA leads
to MAPK activation to
that
MAPK activation
induces that inducesRsk
Rsk phosphorylation. Rskwillphosphorylation. Rsk will ultimately
ultimately phosphorylate CREB. Normal phosphorylate
PC12 cells, CREB.
Cav-1
Normal
PC12 PC12
cells, andcells,
Cav-2 Cav-1
PC12PC12 cells, treated
cells were and Cav-2withPC12 cells were
20 ng/mL NGFtreated
for 0 towith 20 ng/mL
6 h. The NGFstatus
activation for 0 of
to
6 h. Theeffectors
several activation status
of the of several
Ras-MAPK effectorswas
pathway of analyzed
the Ras-MAPK pathway
by Western blotwas
usinganalyzed by directed
antibodies Western
blot using antibodies directed
towards their phosphorylated forms. towards their phosphorylated forms.
treatment of
NGF treatment of normal
normal PC12 cells results
results in a prolonged activation of MAP kinases ERK1 and
ERK2 (extracellular
ERK2 (extracellularsignal-regulated
signal-regulated kinase
kinase 1 & 12),& 2), phosphorylation
with with phosphorylation
peaking peaking
at 30 minattreatment,
30 min
treatment, and detectable up to 2 h after NGF addition (Figure 8A). Rsk2 phosphorylation
and detectable up to 2 h after NGF addition (Figure 8A). Rsk2 phosphorylation is also readily detected is also
readily
and lastsdetected
up to 6 and lasts
h after up to 6of
addition h after
NGF,addition
while CREBof NGF, while CREB phosphorylation
phosphorylation on serine 133 peakson serine 133
at 30 min,
peaks at 30to
decreasing min,
thedecreasing to the limit
limit of detection after of detection after 2 h.
2 h.
Int. J. Mol. Sci. 2017, 18, 693 11 of 25
Int. J. Mol. Sci. 2017, 18, 693 11 of 24

Figure 8. Kinetics of activation of the MAPK pathway in PC12 cells stably-transfected with Cav-1 or
Figure 8. Kinetics of activation of the MAPK pathway in PC12 cells stably-transfected with Cav-1 or
Cav-2. Normal PC12 cells (A); Cav-1 PC12 cells (B) and Cav-2 PC12 cells (C) were exposed for up to
Cav-2. Normal PC1220cells
6 h to NGF ng/mL.(A); Cav-1
Cells werePC12 cells
collected and(B) and Cav-2
proteins PC12 cells
were extracted (C)buffer.
in lysis were Protein
exposed for up
to 6 h to NGF 20 ng/mL.
concentrations in theCells
lysateswere collected and
were determined and40proteins were
µg were used forextracted
Western blot inanalysis.
lysis buffer.
Blots Protein
were in
concentrations probed with phospho-specific
the lysates were determined antibodies
and 40 to µg
MAPK,
wereRsk2,
usedand CREB. Equal
for Western blotloading
analysis.wasBlots were
controlled using a β-tubulin antibody; (D–F) Scans of gels were quantified using the gel analysis
probed with phospho-specific antibodies to MAPK, Rsk2, and CREB. Equal loading was controlled
function of ImageJ software and analyzed as described in the Materials and Methods section; (D) P-
using a β-tubulin antibody; (D–F) Scans of gels were quantified using the gel analysis function of
MAPK; (E) P-Rsk; (F) P-CREB. In all, PC12 cells: Solid line with open circles; Cav-1-PC12 cells: Dotted
ImageJ software
line withand
filledanalyzed as described
squares; Cav-2-PC12: Dashed inline
thewith
Materials and Methods
open triangles. Results are section;
presented (D) P-MAPK;
as the % (E)
P-Rsk; (F)maximal
P-CREB. In all,
response PC12tocells:
in order be ableSolid line multiple
to compare with open circles;(Mean
experiments. Cav-1-PC12 cells:from
± SEM derived Dotted line
with filled24squares;
gels generated in nine independent
Cav-2-PC12: Dashedexperiments
line with performed on Cav-1 Results
open triangles. clones 3, 12,
are16presented
and Cav-2 as the %
clones 5, 11, and 23 (see Supplementary Figure S2) having similar relative expression levels. Statistical
maximal response in order to be able to compare multiple experiments. (Mean ± SEM derived from
significance was ascertained using the two-way ANOVA test with Bonferroni post hoc tests. p values
24 gels generated
are shownin fornine
resultsindependent
obtained with experiments
caveolin clones vs.performed on Cav-1
those obtained clones
with PC12. * p <3, 12,**16
0.01, p <and Cav-2
clones 5, 11, and***23
0.005, p <(see Supplementary Figure S2) having similar relative expression levels. Statistical
0.0005).
significance was ascertained using the two-way ANOVA test with Bonferroni post hoc tests. p values are
Cav-2 PC12 cells exhibit an even longer duration of MAPK, Rsk2, and CREB activation. The
shown for results obtained with caveolin clones vs. those obtained with PC12. * p < 0.01, ** p < 0.005,
phosphorylated forms of the proteins are easily detectable up to 6 h after NGF addition (Figure 8C).
*** p <These
0.0005).
results show that TrkA activation in Cav-2 PC12 cells is accompanied by the lengthening of the
duration of MAPK pathway activation. Extended MAPK activation has been proposed to be critical
Cav-2inPC12
mediating the differentiation effects of NGF in PC12 cells [60], this could provide an explanation
cells exhibit an even longer duration of MAPK, Rsk2, and CREB activation. The
for the potentiating effects of Cav-2 expression on NGF-induced differentiation.
phosphorylated forms of the
In Cav-1 PC12 cells,proteins
Erk1 and are
Erk2easily detectable
are activated up to very
with kinetics 6 h after
similarNGF
to thataddition (Figure 8C).
observed in
These results
normalshow
PC12that
cellsTrkA
(Figureactivation
8B). RSK2 isinalso
Cav-2 PC12
activated, cells isphosphorylation
although accompanieddoes by not
thelast
lengthening
as long of the
duration ofasMAPK
that observed
pathway in normal PC12 cells.
activation. CREB phosphorylation
Extended MAPK activation on Serine
has133been
was proposed
barely detectable.
to be critical in
mediatingThis
theresult shows that the activation of CREB is inhibited in Cav-1 PC12 cells. Thus, it appears that
differentiation effects of NGF in PC12 cells [60], this could provide an explanation for
Cav-1 expression blocks CREB activation while Cav-2 expression sustains it.
the potentiating effects of Cav-2 expression on NGF-induced differentiation.
In Cav-1 PC12
2.7. Effect cells,and
of Cav-1 Erk1
Cav-2and Erk2 are
Expression activated
on TrkA with kinetics very similar to that observed in
Effector Localization
normal PC12 cellsNormal activation of CREB in response to NGF is achieved viaphosphorylation
(Figure 8B). RSK2 is also activated, although does 133
phosphorylation of serine notbylast as long
as that observed in normal
Rsk2 [61,62]. FollowingPC12 cells.pRsk2
activation, CREB phosphorylation
is translocated on Serine
into the nucleus where133 was barely detectable.
it phosphorylates
This resultCREB
shows [62].that
Sincethe
Rsk2 is phosphorylated
activation of CREB in Cav-1 overexpressing
is inhibited PC12,PC12
in Cav-1 we investigated if theitlack
cells. Thus, appears that
Cav-1 expression blocks CREB activation while Cav-2 expression sustains it.

2.7. Effect of Cav-1 and Cav-2 Expression on TrkA Effector Localization

Normal activation of CREB in response to NGF is achieved via phosphorylation of serine 133 by
Rsk2 [61,62]. Following activation, pRsk2 is translocated into the nucleus where it phosphorylates
CREB [62]. Since Rsk2 is phosphorylated in Cav-1 overexpressing PC12, we investigated if the lack
of CREB phosphorylation was due to impaired pRsk translocation. Subcellular localization of pRsk2,
CREB and pCREB was studied by confocal immunofluorescence microscopy in the caveolin-expressing
Int. J. Mol. Sci. 2017, 18, 693 12 of 25

Int. J. Mol. Sci. 2017, 18, 693 12 of 24

cells. Figure 9A,B show that pCREB was detected in the nucleus of normal PC12 cells and Cav-2 PC12
cells withofaCREB
greater phosphorylation was due to impaired pRsk translocation. Subcellular localization of pRsk2,
level of activation in Cav-2 PC12 cells. In the nucleus of Cav-1 PC12 cells, the
CREB and pCREB was studied by confocal immunofluorescence microscopy in the caveolin-
level of CREB activation
expressing was9A,B
cells. Figure essentially
show thatcompletely inhibited.
pCREB was detected Furthermore,
in the nucleus of normal in normal
PC12 cellsPC12
and cells and
in Cav-2 PC12
Cav-2 PC12cellscells
exposed to 20 ng/mL
with a greater NGF for
level of activation 30 min,
in Cav-2 PC12pRsk2
cells. Inisthe
mainly
nucleuslocated in the nucleus.
of Cav-1 PC12
cells,inthe
By contrast, level ofPC12
Cav-1 CREB cells
activation was essentially
exposed to NGF, completely
the levelinhibited.
of pRsk2 Furthermore, in normalcompartment
in the nuclear PC12 is
cells and in Cav-2 PC12 cells exposed to 20 ng/mL NGF for 30 min, pRsk2 is mainly located in the
drastically diminished. Moreover, accumulation of pRsk2 can be seen in isolated foci at the periphery
nucleus. By contrast, in Cav-1 PC12 cells exposed to NGF, the level of pRsk2 in the nuclear
of the cell. Thus theisabsence
compartment drasticallyofdiminished.
enhanced nuclearaccumulation
Moreover, translocation of pRsk2
of pRsk2 in Cav-1
can be seen PC12
in isolated foci cells may
explain whyat theCREB is not
periphery of phosphorylated in theofnucleus
the cell. Thus the absence enhancedof thesetranslocation
nuclear cells. of pRsk2 in Cav-1
PC12 cells may explain why CREB is not phosphorylated in the nucleus of these cells.

Effect
Figure 9.Figure of ofCav-1,
9. Effect Cav-2
Cav-1, Cav-2 and and Cav-1
Cav-1 S80V S80V on
expression expression
TrkA effectoron TrkA effector
localization. (A) Normal localization.
PC12 cells, Cav-1 PC12 cells Cav-2 PC12, and Cav-1 S80V PC12 cells were
(A) Normal PC12 cells, Cav-1 PC12 cells Cav-2 PC12, and Cav-1 S80V PC12 cells were plated on plated on collagen/poly-
lysine coated coverslips and exposed to 20 ng/mL NGF for 30 min after 17 h of serum deprivation.
collagen/poly-lysine coated coverslips and exposed to 20 ng/mL NGF for 30 min after 17 h of serum
Cells were fixed and simultaneously stained with an anti-pRsk2 antibody (cyan), an anti-CREB
deprivation. Cells were fixed and simultaneously stained with an anti-pRsk2 antibody (cyan), an
antibody (green) and an anti-pCREB antibody (red). Cells were mounted in mounting medium
anti-CREB antibody (green)
containing DAPI to visualizeand thean nucleus;
anti-pCREB antibody of
(B) Quantification (red).
CREB Cells were mounted
phosphorylation in mounting
level in the
nucleus. For each
medium containing DAPI single cell, quantitation
to visualize of fluorescence
the nucleus; representative ofof
(B) Quantification pCREB
CREB and CREB in the
phosphorylation level
nucleus was assessed using ImageJ software. Level of CREB phosphorylation was evaluated by
in the nucleus. For each single cell, quantitation of fluorescence representative of pCREB and CREB
dividing pCREB fluorescence by CREB fluorescence. Results are represented as percent of CREB
in the nucleus was assessed using ImageJ software. Level of CREB phosphorylation was evaluated
phosphorylation observed with CREB phosphorylation in normal PC12 cells considered as 100
by dividing pCREB
percent. (Mean fluorescence
of 20 to 70 cellsby per
CREB fluorescence.
conditions Resultssignificance
± SEM. Statistical are represented
vs. PC12as percent
cells, was of CREB
phosphorylation
ascertainedobserved
using thewith CREB
unpaired, phosphorylation
two-tail Student’s t-Test.in
** normal PC12
p < 0.01; *** cells considered as 100 percent.
p < 0.001.)
(Mean of 20 to 70 cells per conditions ± SEM. Statistical significance vs. PC12 cells, was ascertained
using the unpaired, two-tail Student’s t-Test. ** p < 0.01; *** p < 0.001.)

Figure 10 shows the cytolocalization of Cav-1, Cav-1 S80V and pRSK2 after 30 min exposure of
normal, Cav-1 and Cav-1 S80V PC12 cells to 20 ng/mL NGF. Cav-1 was predominantly located at the
periphery of the Cav-1 PC12 cells, whereas Cav-1 S80V was essentially located intracellularly.
Int. J. Mol. Sci. 2017, 18, 693 13 of 24

Figure 10 shows the cytolocalization of Cav-1, Cav-1 S80V and pRSK2 after 30 min exposure of
Int. J. Mol. Sci. 2017, 18, 693
normal, Cav-1 and Cav-1 S80V PC12 cells to 20 ng/mL NGF. Cav-1 was predominantly located 13 at of
the25

periphery of the Cav-1 PC12 cells, whereas Cav-1 S80V was essentially located intracellularly.

Figure
Figure 10.
10. Cav-1,
Cav-1,Cav-1
Cav-1S80V,
S80V, and
and phospo-Rsk2
phospo-Rsk2 localization
localization in
in normal,
normal, Cav-1
Cav-1 and
and Cav-1
Cav-1 S80V
S80V PC12
PC12
cells.
cells. Normal PC12 cells, Cav-1, and Cav-1 S80V PC12 cells were plated on collagen/poly-lysinecoated
Normal PC12 cells, Cav-1, and Cav-1 S80V PC12 cells were plated on collagen/poly-lysine coated
coverslips
coverslips and
and exposed
exposed toto 20
20 ng/mL
ng/mLNGFNGFfor
for30
30min
minafter
after17
17hhofofserum
serumdeprivation.
deprivation.Cells
Cellswere
werefixed
fixed
and simultaneously stained with an anti-pRsk2 antibody (cyan) and an anti-Cav-1 antibody
and simultaneously stained with an anti-pRsk2 antibody (cyan) and an anti-Cav-1 antibody (green). (green).

2.8. Effect of Cav-1 Point Mutations on the NGF Signaling Pathway

2.8. Effect of Cav-1 Point Mutations on the NGF Signaling Pathway
Paracrine/autocrine activation of NGF signaling has been implicated in tumor progression, cell
Paracrine/autocrine activation of NGF signaling has been implicated in tumor progression, cell
survival and metastasis and deletion or mutation of Cav-1 have been reported for certain cancers,
survival and metastasis and deletion or mutation of Cav-1 have been reported for certain cancers,
notably breast cancer [63–67] while overexpression has been observed in certain prostate cancer lines
notably breast cancer [63–67] while overexpression has been observed in certain prostate cancer
[26,68]. Mutations of specific codons, e.g., those coding for Pro132 and Ser 80, have been reported to
lines [26,68]. Mutations of specific codons, e.g., those coding for Pro132 and Ser 80, have been reported
potentially override or inactivate the growth inhibitory activity of Cav-1 [24,69,70].
to potentially override or inactivate the growth inhibitory activity of Cav-1 [24,69,70].
PC12 cells were stably-transfected with Cav-1 S80V (non-phosphorylatable form) and assayed
PC12 cells were stably-transfected with Cav-1 S80V (non-phosphorylatable form) and assayed
for transgene expression. Individual clones expressing Cav-1 S80V were selected, and the impact on
for transgene expression. Individual clones expressing Cav-1 S80V were selected, and the impact on
specific parameters evaluated as described above with cells expressing wild-type caveolins. Figure
specific parameters evaluated as described above with cells expressing wild-type caveolins. Figure 1A,B
1A,B show that, Cav-1 S80V PC12 cells exposed to NGF developed a more extensive and denser
show that, Cav-1 S80V PC12 cells exposed to NGF developed a more extensive and denser neurite
neurite network than normal PC12 cells. This enhanced morphological differentiation was observed
network than normal PC12 cells. This enhanced morphological differentiation was observed even
even at low concentrations of NGF. The NGF-induced morphological differentiation of Cav-1 S80V
at low concentrations of NGF. The NGF-induced morphological differentiation of Cav-1 S80V PC12
PC12 cells appears to be potentiated in comparison to normal PC12 cell response, as was observed
cells appears to be potentiated in comparison to normal PC12 cell response, as was observed with
with the differentiation of Cav-2 PC12 cells (Figure 1). Figure 2A shows that, following NGF exposure,
the differentiation of Cav-2 PC12 cells (Figure 1). Figure 2A shows that, following NGF exposure,
Cav-1 S80V PC12 cells and normal PC12 cells stop proliferating with similar kinetics. Thus,
Cav-1 S80V PC12 cells and normal PC12 cells stop proliferating with similar kinetics. Thus, expression
expression of Cav-1 S80V does not result in the inhibition of the anti-mitogenic effect nor of the
of Cav-1 S80V does not result in the inhibition of the anti-mitogenic effect nor of the differentiation
differentiation response that are observed with Cav-1 expression. Furthermore TrkA internalization
response that are observed with Cav-1 expression. Furthermore TrkA internalization is not affected
is not affected by expression of Cav-1 S80V, in contrast to that of Cav-1 (Figure 3). Finally, in Cav-1
by expression of Cav-1 S80V, in contrast to that of Cav-1 (Figure 3). Finally, in Cav-1 PC12 cells,
PC12 cells, the reduction of nuclear pRsk2, concomitant with the accumulation of pRsk2 at the cell
the reduction of nuclear pRsk2, concomitant with the accumulation of pRsk2 at the cell periphery,
periphery, is associated with an almost total inhibition of CREB phosphorylation. In contrast, NGF
is associated with an almost total inhibition of CREB phosphorylation. In contrast, NGF treatment
treatment of Cav-1 S80V PC12 cells results in nuclear localization of pRsk2 along with a high level of
of Cav-1 S80V PC12 cells results in nuclear localization of pRsk2 along with a high level of pCREB,
pCREB, superior to that observed in NGF treated normal PC12 cells (Figure 8A,B).
superior to that observed in NGF treated normal PC12 cells (Figure 8A,B).
Int. J. Mol. Sci. 2017, 18, 693 14 of 25

Taken together, these results show, that while expression of Cav-1 inhibits TrkA trafficking and
NGF signaling, the Cav-1 S80V mutant does not, and actually potentiates the NGF response.

3. Discussion
The aim of this study was to gain insight into the mechanisms underlying the effects of membrane
microdomain components Cav-1 and Cav-2 on growth factor signaling using the NGF receptors TrkA
and p75NTR as model.
A “ying and yang” dichotomy between Cav-1 and Cav-2 was observed, with Cav-1 inhibiting
and Cav-2 potentiating the response to NGF. Cav-1 expression impaired exit from lipid rafts and
internalization of NGF receptors, TrkA and p75NTR , without, however, abrogating TrkA short-term
activation, nor downstream effector activation up to and including Rsk2. Remarkably, Cav-1 expression
was associated with accumulation of activated Rsk2 at the plasma membrane and subsequent inhibition
of CREB phosphorylation. These results indicate that Cav-1 is retaining the TrkA receptor signaling
complex up to and including Rsk2 in the activated state at the plasma membrane, thereby preventing
its nuclear translocation and phosphorylation of CREB.
By contrast, PC12 cells expressing Cav-2, displayed normal kinetics of cell-cycle arrest when
exposed to NGF, while morphological differentiation was potentiated. In Cav-2 PC12 cells, NGF
receptor exit from lipid rafts and internalization, as well as nuclear localization of pRsk2, were similar
to that observed in PC12 cells. However, phosphorylated forms of TrkA effectors MAPK, Rsk2
and CREB, exhibited higher levels and/or longer activation kinetics. Taken together, these results
demonstrate differential modulation of NGF signaling events by Cav-1 and Cav-2.

3.1. NGF-Induced Exit of TrkA and p75NTR from Lipid Rafts

Numerous growth factor receptors have been observed to move in to or out of membrane
microdomains [71,72], including neurotrophic factor receptors [73,74]. Cav-1 has been reported to play
a role, independent of clathrin, in the internalization and sorting of several growth factor receptors
including fibroblast growth factor receptor [75] and insulin receptor [76]. Here we show that NGF
induces TrkA and p75NTR exit from lipid rafts in PC12 cells. It has been reported that TrkA and
TrkB reside outside lipid rafts prior to neurotrophic factor stimulation, then concentrate in lipid rafts
consecutive to treatment with their respective ligands [77–79]. The case is similar for p75NTR in
sympathetic neurons [13]. By contrast, a number of studies have provided evidence that, in the absence
of treatment, TrkA is enriched in lipid rafts [11,12,14,15,17,80], as observed herein.
The use of different methods for lipid raft isolation could explain the discrepancies. The studies
presented herein offer insight for investigating the role of the caveolins in other experimental
paradigms, of which growth factor signaling in particular, thereby contributing to our growing
understanding of their diverse actions. Clearly, the very definition of growth factor binding-initiated
translocation into, or out of lipid rafts, is dependent on the extraction protocol. While we, and
others [11,12,14,17,80] used a detergent-independent method, either on post-mortem tissue or in
cultured cells [13,77–79], used a Triton X100-dependent method, which has been shown to induce
the formation of non-physiological structures [81]. This discrepancy notwithstanding, the common
conclusion of these observations is that there is ligand-induced movement of the receptors between
different membrane compartments. This underlines an interesting paradox that warrants clarification,
which would offer insight on the microenvironment of the receptors in the different cellular membranes.
The use of microscopy, and high-resolution microscopy, in particular [76,82,83], will help to substantiate
the previous observations, and validate which extraction protocol(s) reflect what is actually occurring
in and on the cell.

3.2. Cav-1 Mode of Action

Expression of Cav-1 in DRG, attenuated the differentiation response. The results presented
herein confirm that, in PC12 cells, expression of Cav-1 inhibits NGF-induced morphological
Int. J. Mol. Sci. 2017, 18, 693 15 of 25

differentiation [14,16]. This is due therefore, to inhibition of NGF-induced TrkA and p75NTR exit
from lipid rafts by Cav-1. It has been proposed by Lajoie and Nabi 2007 that Cav-1 negatively regulates
endocytosis either by stabilizing raft invagination at the cell surface or by sequestering key structural
components, notably including cholesterol and dynamin [84,85], independently of the presence of
caveolae [86]. The fact that pRsk2 accumulates at the periphery of the cell following a pattern closely
resembling that of Cav-1, and TrkA (as illustrated in the dynamic experiments of Figure 5) suggests
that Cav-1 is indeed sequestering the TrkA signaling complex at the cytoplasmic membrane in these
cells. This effect is dependent on the presence of Serine 80, as demonstrated by the results obtained
with the S80V mutant. The limited studies on this residue suggest a key role in regulating the binding
of free cholesterol in cell membranes [87]. Whether this has an impact on Cav-1’s effect on TrkA
trafficking remains to be determined.
Cav-1 has been reported to directly inhibit EGFR tyrosine kinase activity in vitro [88]. The results
presented herein indicate that Cav-1 expression in PC12 cells results in the prevention of the
anti-proliferative response, at least in part via the prevention of the induction of p21Cip1/Waf1 . However,
the NGF-TrkA, Ras, Raf, MAPK-dependent signaling pathway is not inhibited per se since TrkA, MAPK,
and Rsk2 are indeed activated with similar kinetics in PC12 clonal lines stably-expressing Cav-1 as
compared to PC12 cells in response to NGF. CREB phosphorylation, the triggering event essential for
propagation of the anti-proliferative and differentiation responses, cannot occur since the activated
Rsk2 is retained at the cell membrane.

3.3. MC192 Induced NGF Receptor Immobilization in Rafts

MC192 treatment mimics Cav-1 effects on internalization of TrkA [58]. MC192 binds to the
extracellular domain of p75NTR , and without preventing the binding of NGF, attenuates NGF
signaling via TrkA [57,89]. Application of MC192 to adult DRG resulted in a slight decrease in TrkA
phosphorylation (30 min and 24 h) and yet was shown to significantly attenuate, if not block, neurite
outgrowth [90]. The results presented herein offer novel insight into the mechanism via which MC192
is acting on NGF signaling. Namely, they indicate that MC192 prevents NGF-induced exit of both
p75NTR and TrkA from lipid rafts. This result suggests that the conformational change provoked (or
prevented) by the binding of MC192 to p75NTR directly impacts upon TrkA conformation, preventing
p75NTR ’s NGF-dependent exit from lipid rafts and, thus, internalization. The sequestration in lipid
rafts therefore affords an initial explanation of the previously observed reduction of internalization
of TrkA [58]. Further, this observation offers a novel means by which to study the dynamics of the
intimate interactions between these two receptors that have been so elusive to the field.

3.4. Effect of Cav-2 on the NGF Signaling Pathway

Cav-2 expression in PC12 cells resulted in enhanced morphological differentiation. Very little is
known about Cav-2, which has often been considered an accessory protein [91–93]. However Cav-2
has been shown to be directly involved in endocytosis of bacteria and apical trafficking of lipids [94,95].
Cav-2 is upregulated in esophageal and bladder carcinomas [96,97], which is consistent with the idea
that Cav-2 could potentiate the MAPK pathway. Here, the potentiating effect of Cav-2 is associated
with an increase and/or a longer duration of downstream effector activation as compared to normal
PC12 cells.
Alternatively Cav-2 could promote TrkA signaling in a manner similar to that which has been
observed for the insulin receptor (IR). Notably, Cav-2 phosphorylation prolonged IR activation by
preventing its interaction with an inhibitory protein, SOCS-3 [98].
However the bulk of Cav-2 is not targeted to the plasma membrane and is localized in the Golgi
apparatus [99]. This observation suggests that Cav-2 modulation of the NGF signaling pathway may
not be mediated by the direct interaction between Cav-2 and TrkA at the plasma membrane. One
possible mechanism by which Cav-2 could trigger its effects is via retaining a negative regulator of
TrkA signaling in the Golgi apparatus, such as a phosphatase [100].
Int. J. Mol. Sci. 2017, 18, 693 16 of 25

4. Materials and Methods

4.1. Reagents
Cell culture media is from Thermo Fisher Scientific (Courtaboeuf, France). Fetal bovine serum
and horse serum were from Sigma (St. Quentin Fallavier, France) and GE-PAA (Velizy-Villacoublay,
France), respectively. Transfection reagent lipofectamine 2000 is from Thermo Fisher. Dual luciferase
kits are from Promega (Madison, WI, USA). ECL chemiluminescence system detection system
is from GE-Amersham Pharmacia Biotech (Velizy-Villacoublay, France). ProLong Gold antifade
reagent is from Invitrogen (Thermo Fischer Scientific, Courtabeuf, France) and Vectashield with
DAPI is from Abcys (Courtaboeuf, France). NGF 2.5S Grade II from mouse submaxillary glands
was purchased from Alomone Labs (Jerusalem, Israel). Anti-Histone H1 antibody is from Upstate
Biotechnology (Lake Placid, NY, USA). Anti-TrkA extracellular domain (RTA) [56] was a kind gift of
Louis F. Reichardt (University of California, San Francisco, CA, USA). Anti-p75NTR carboxy-terminus
serum is a kind gift of Moses. V. Chao (New York University Medical Center, New York, NY, USA).
Anti-p75NTR extracellular domain mAb MC192 is from Merck Millipore—Chemicon International
(Fontenay sous Bois, France). Anti p21 mAb CP36 is a kind gift of Wade Harper (Baylor College of
Medicine, Houston, TX, USA). Anti-Trk C-14, anti-phosphotyrosine (PY99), anti-pRsk2, and anti-Cav-1
antibodies are from Santa Cruz Biotechnology (Dallas, TX, USA). Anti-phospho-MAPK antibodies are
from Promega. Anti-pCREB antibody is from Ozyme (Saint Quentin Yvelines, France). Anti-Cav-2
mAb is from Transduction Laboratories (Le Pont de Claix, France). Anti-β-tubulin mAb is from
Sigma. Protein A Sepharose 4 Fast Flow is from Amersham Pharmacia. HRPO-linked anti-rabbit or
anti-mouse IgG secondary antibody where purchased from GE Healthcare (Velizy-Villacoublay, France).
Dylight Fluo-linked anti-rabbit or anti-mouse IgG secondary antibodies are from ThermoFisher
Scientific. Alexa-Fluo linked anti-rabbit, anti-mouse and anti-goat antibodies are from Invitrogen.
K252a is from Merck-Calbiochem (Fontenay sous Bois). The siRNA against caveolin-1 (sc-106996);
caveolin-2 (sc-270431) and FITC-conjugated scrambled siRNA (sc-36869) are from Santa Cruz. siRNA
transfection reagent lipofectamine RNAiMAX is from Life Technologies (Thermo Fisher Scientific,
Courtaboeuf, France).

4.2. Cell Culture and Transfection

PC12 cells were grown as previously described [42]. cDNAs of human caveolin 1 (IMAGE
#488533) and human caveolin 2 (IMAGE #491497) were purchased from Research Genetics (Huntsville,
AL, USA). Caveolin 1 and 2 cDNAs were subcloned in the pcDNA3.1 (+) vector (Clontech) and correct
cloning was verified by sequencing. PC12-Cav-1 and PC12-Cav-2 monoclonal populations were
obtained by calcium-phosphate transfection of PC12 cells with pcDNA3.1-Cav-1 or pcDNA3.1-Cav-2
constructs, followed by an initial selection of 4 weeks in the presence of 0.4 mg/mL of G418 (Life
Technologies). Antibiotic-resistant colonies were assayed for transgene expression by immunoblot
analysis. Taking pcDNA3.1-Cav-1 plasmid as a template pcDNA3.1-Cav-1 S80V mutant was generated
using QuickChange XL Site directed Mutagenesis Kit Agilent-Stratagene (Les Ulis, France) according to
the manufacturer’s instruction. Primers were purchased from MWG-Biotech (Ebersberg, Germany) AG:
S80V mutant: sense: 50 -CCA GAA GGG ACA CAC GTG TTT GAC GGC ATT TGG AAG GCC AGC-30 ,
anti-sense: 50 -GCT GGC CTT CCA AAT GCC GTC AAA CAC GTG TGT CCC TTC TGG-30 (codon in
bold is the position 80). PC12-Cav-1 S80V monoclonal populations were obtained by lipofectamine
2000 transfection of PC12 cells with pcDNA3.1-Cav-1 S80V constructs, followed by an initial selection
of four weeks in the presence of 0.8 mg/mL of G418 (Life Technologies). Antibiotic-resistant colonies
were assayed for transgene expression by immunoblot analysis.

4.3. siRNA Transfection

2 × 104 cells were seeded per well onto 24-well plates in culture medium 8–12 h before transfection.
The transfection of siRNA at a final concentration of 30 nM was performed using Lipofectamine
Int. J. Mol. Sci. 2017, 18, 693 17 of 25

RNAiMAX according to the manufacturer’s instructions. FITC-conjugated scrambled siRNA was used
as a non-targeting control siRNA. Medium (without serum) was refreshed 8–12 h after transfection,
cells were then maintained in DMEM medium supplemented with 0 or 20 ng/mL NGF for 72 h before
imaging. The neurite outgrowth was quantified using NeuronJ plug-in for ImageJ (National Institutes
of Health, Bethesda, MD, USA). Cells were considered “with neuritis” when neurites were ≥3 cell
diameters. Statistical significance was tested with the two-tail unpaired Student’s t-test.

4.4. DRG Neuron Culture, Transfection, and Quantification

All procedures were performed in accordance with French and European legislation on
animal experimentation. Primary DRG neurons were prepared from E14.5 mouse embryos and
electroporated as previously described [37] with pEGFP (Clontech, Saint-Germain-en-Laye, France)
(0.5 µg) and pCDNA3.1-Cav-1 fused to RFP (gift from Ari Helenius, Addgene Plasmid 14434) [38]
or pCDNA3.1-RFP (1 µg) endotoxin free plasmids (XtraMaxi, Nucleobond, Macherey Nagel, Hoerdt,
France). Control RFP was derived from the Cav-1-RFP plasmid. The Cav-1 coding sequence was
removed by BamHI and Not1 digestion and plasmid ligation was performed after Kleenow fill-in.
Because of the low number of cells required, both control and Cav-1 electroporations could be
performed on the same batch of dissociated cells.
Electroporated cells were plated on poly-L-Polylysine and laminin coated coverslips (PLL:
50 µg/mL; laminin: 10 µg/µL, Sigma. 1 h after plating, medium (F12 + 10% FCS) was complemented
with 50 ng/mL NGF (N0513, Sigma) and antibiotics (1% final, 15140, Gibco, Courtaboeuf, France).
10 µM of AraC was added the next morning (C1768, Sigma).
Control and caveolin-1 cultures from the same dissociation batch were fixed at different time
points in 2% paraformaldehyde at 4 ◦ C and neurite growth was analyzed. Four montages of nine
images were collected in different regions of the coverslips to minimize bias due to different cell density.
Number of transfected neurons per image was calculated and neurite length were quantified from the
GFP channel in ImageJ (National Institutes of Health, Bethesda, MD, USA; http://imagej.nih.gov/ij)
using NeuronJ plugin [101]. Statistical significance was tested with the two-tail unpaired Student’s
t-test.

4.5. Western Blot Analysis

To obtain lysates, cells were washed once and collected in ice cold phosphate-buffered saline
(PBS). Pelleted cells were resuspended in lysis buffer (20 mM tris-HCl, pH 8.0; 137 mM NaCl; 2 mM
EDTA; 10% glycerol; 1% Nonidet P-40; 20 µM leupeptin; 1 mM sodium vanadate; 1 mM Pefabloc;
0.15 U/mL aprotinin; 1 mM ß-glycerophosphate; 3 mM sodium-fluoride). After 30 min incubation on a
rotating wheel at 4 ◦ C, the extracts were clarified by centrifugation at 12,000× g for 20 min. The protein
concentration of the supernatants was quantitated using the DC (detergent compatible) protein assay
(Bio-Rad, Marnes-la-Coquette, France). Next, 5X-SDS-PAGE sample buffer was added to the lysates
prior to boiling for 5 min. Proteins were separated by SDS-PAGE and transferred to nitrocellulose
membranes. Membranes were blocked in TBST (25 mM Tris-HCl pH 7.4, 137 mM NaCl, 3 mM KCl,
0.1% Tween-20) containing 5% (w/v) dried milk or 5% (w/v) bovine serum albumin (BSA) (sigma)
depending on the primary antibody, incubated 1 h at room temperature with primary antibody diluted
in TBST with 5% BSA. After two washes with TBST, membranes were incubated 1 h with horseradish
peroxidase-conjugated secondary antibodies, washed twice with TBST and immunoreactive proteins
were visualized using chemiluminescence.
Quantification of the scanned gels was performed using the Gel analysis routine of ImageJ
either with specification of individual lanes or selection of the entire row as described in the online
documentation. The values obtained were normalized for those from the corresponding tubulin
loading control and, subsequently, to the maximum signal in a given series, in order to be able
to compare results obtained from multiple experiments. Statistical significance between selected
conditions was ascertained with the unpaired, two-tail Student’s t-test.
Int. J. Mol. Sci. 2017, 18, 693 18 of 25

Alternatively, to obtain quantitative results in some experiments, protein expression was evaluated
by SDS-PAGE followed with a western blot analysis using the Odyssey. In this case, membranes were
blocked in TBS (25 mM Tris-HCl pH 7.4, 137 mM NaCl, 3 mM KCl) containing 5% (w/v) BSA, incubated
1 h at room temperature with primary antibody diluted in TBST with 5% BSA. After four washes
with TBST, membranes were incubated 1 h with Dylight Fluo-linked secondary antibody allowing
quantitative infrared fluorescence detection using Odyssey imaging system Odyssey (ODY-1092,
ScienceTec, Villebon-sur-Yvette, France).

4.6. Luciferase Assays

Wild type PC12 cells, Cav-1 PC12 cells and Cav-2 PC12 cells were spread on collagen/poly
L-lysine-coated 6 well dishes [58] and transfected with lipofectamine 2000 with 2 µg of p21P93SLuc
reporter as previously described [45,102] and 0.1 µg of the reporter pEGFP-C2 or alternatively with
0.1 µg of pCMV H-RasV12. 24 h after transfection, cells were transferred to 96-well plates, and treated
or not with 50 ng/mL of NGF. 48 h later, cells were lysed and assayed for luciferase activity using the
Dual Luciferase kit (Promega).

4.7. Immunoprecipitation
Following NGF treatment, cells were collected and lysed. Cleared lysates (1 mg protein per
immunoprecipitate) were then incubated on a rotating wheel for 2 h at 4 ◦ C with 2 µg of Trk C-14
antibody. Protein A Sepharose beads were added to bind the antibodies; after 2 h at 4 ◦ C beads were
washed for times with lysis buffer and proteins were eluted by boiling for 5 min in sample buffer.

4.8. Isolation of Lipid Rafts

Lipid rafts (LR) were isolated essentially as described [103]. Briefly, PC12 cells from a 162 cm2 cell
culture flask grown to 60–70% confluence were collected in ice-cold PBS and resuspended in 1.5 mL of
Na2 CO3 0.5 M pH 11.0. Homogenization was carried out using a sonicator (30 three-second bursts on
ice; Vibra Cells, Sonics & Materials, Newtown, CT, USA). The homogenate (6 mg protein/1.5 mL) was
then adjusted to 45% sucrose by addition of 1.5 mL of 90% sucrose prepared in H2 O and loaded in a
Beckman ultracentrifuge tube under a 5–35% discontinuous sucrose gradient (3 mL of 5% sucrose/6
mL of 35% sucrose; both in 250 mM Na2 CO3 pH 11.0). After centrifugation at 39,000 rpm for 14 h in a
SW41 rotor (Beckman, Villepinte, France), lipid rafts were apparent as a light-scattering band confined
to the 5–35% sucrose interface. Twelve fraction of 1 mL were collected by pipetting from the top of the
gradient 20 µL were used for SDS-PAGE. The lipid rafts were present in fractions 3 and 4.

4.9. Confocal Immunofluorescence Microscopy

PC12 cells were spread on collagen/poly L-lysine-coated cover slips [58]. Cells were rinsed
twice with PBS, fixed 30 min in PBS 3.7% formaldehyde and permeabilized one minute in PBS
0.5% Triton X-100. After washing with PBS, cells were blocked with PBS 0.5% BSA for 30 min and
incubated 1 h with primary antibodies diluted in blocking buffer. Cells were washed three times
in PBS and incubated 1 h with secondary antibodies conjugated to Alexa 488, Alexa 555, or Alexa
647. Cells were washed again three times in PBS and coverslips were mounted with ProLong Gold
antifade or Vectashield with DAPI reagent. Scanning fluorescence images were acquired using the
DM6000-confocal unit coupled to a Leica (Nanterre, France) spectral confocal TCS SP5 AOBS with
a HCX Plan APO 63/1.2 W Corr oil-immersion objective. Data acquisition was performed with
Leica LAS AF SP5 software. Quantitation of the fluorescence was assessed using ImageJ software.
For the study of TrkA internalization, cells were transiently transfected with a TrkA-EGFP fusion
protein expression vector [55] using lipofectamine 2000, and immunostaining was performed 48 h
post-transfection. Cells were preincubated for 15 min at 4 ◦ C with RTA antibody (1/500 dilution). Cells
were then replaced at 37 ◦ C for 20 min prior to standard fixation and immunostaining procedure as
described above. Scanning fluorescence images were acquired using the MRC1000-confocal laser unit
Int. J. Mol. Sci. 2017, 18, 693 19 of 25

(Bio-Rad Labs, Hercules, CA, USA) coupled to a Zeiss (Marly-le-Roi, France) Axioplan Microscope
equipped with a Zeiss Plan-Apochromat 63X/1.4 oil-immersion objective, an LSM 510 camera. Data
acquisition was performed with LSM510 software.

4.10. Statistical Analysis

Averages and standard errors of the mean were calculated and analyzed for statistical significance
vs. controls, using the unpaired, two-tail Student’s t-test (Figures 1H, 2B,C, 4B, 6B, 7B and 9B).
For Figure 3C, the regular two-way ANOVA was applied twice using Prism© 5 for Windows® , version
5.01. First ± NGF was compared in each cell type followed by the Bonferroni correction post hoc test (*
p values). The second pass compared—NGF between cell types and + NGF between cell types, again
followed by the Bonferroni correction post hoc test (# p values). For Figure 8D–F, the two-way ANOVA
test with the Bonferroni post hoc test was applied via Microsoft® Excel© . For Mac 2011, version 14.5.5
(150821) with StatPlusMac® (AnalystSoft, Walnut, CA, USA).

5. Conclusions
The data presented herein provide evidence for a differential role of Cav-1 and Cav-2 on NGF
signaling. Caveolins are key components of NGF receptor microenvironment that play an important
role in the outcome of NGF signaling. Cav-1 inhibits NGF-induced cell responses acting on the
trafficking of p75 NTR , TrkA, and its downstream effectors by sequestration of key signaling molecules
at the cell membrane, resulting in the prevention of their nuclear translocation and phosphorylation of
a key transcription factor CREB. Cav-1 S80V, a non-phosphorylatable form of Cav-1, does not present
the same inhibitory effects as Cav-1 on TrkA trafficking and downstream signaling, suggesting that
phosphorylation of this residue is key to this effect. It contributes to understanding the molecular
impact of specific Cav-1 mutations in certain cancers where TrkA, p75NTR , and NGF are involved in
the tumorigenic response, while raising the question of the potential impact on trafficking of other
receptors and channels. Cav-2 has the opposite effect to Cav-1, enhancing the cellular response to NGF.
This highlights the opportunity for gaining a better understanding of the impact of major lipid raft
components on NGF receptor trafficking and subsequent NGF signaling in the context of cancer and
neuronal differentiation.

Supplementary Materials: Supplementary materials can be found at www.mdpi.com/1422-0067/18/4/693/s1.

Acknowledgments: We wish to thank Moses V. Chao and Louis F. Reichardt for their generous gifts of antibodies
towards p75NTR and TrkA, respectively and Rudy Marty for assistance with the statistical analysis. This work was
supported by MIRA grants from the Rhone Alps Region of France, the Agence National de la Recherche (ANR)
RIB and the CNRS (International program for Scientific Cooperation, “PICS”) to Brian B. Rudkin and from East
China Normal University (ECNU) to Chonggang Yuan. Personnel support was as follows: Ministry of Education
of China—YinDi Ding, Lingli Yu; Ministry of Foreign Affairs of France, the Rhone Alps Region—YinDi Ding,
Ambre Spencer, Lingli Yu, Ji Ma and the CNRS—David Cluet; East China Normal University—YinDi Ding, Lingli
Yu, Ambre Spencer; the ARC (Association pour la Recherche contre le Cancer)—Ambre Spencer, Vincent Guili,
Jérôme Jullien; Ministry of Higher Education & Research of France—Vincent Guili, Jérôme Jullien; The Ligue
Nationale Contre le Cancer—David Koubi.
Author Contributions: The authors participated in the design and performed and analyzed the experiments
depicted in the figures as follows: Ambre Spencer 2–7, 9, 10, T1; Lingli Yu 1–4, 6, 10, T1, S1; Vincent Guili 2–10,
T1, S2; David Koubi 2–5, 9, 10, T1; YinDi Ding 4; Ji Ma 3, 4; Jérôme Jullien 2, 3, 5, T1; Florie Reynaud, Julien Falk
1; Emmanuel Gauthier offered technical assistance for plasmid preparation, David Cluet for PC12 cell culture
and manipulation, lipid raft preparation, and general laboratory operations. Valérie Castellani coordinated
experiments performed at the CGphiMC on DRG; Chonggang Yuan conceived and coordinated experiments
performed at East China Normal University on DRG, lipid rafts, and differentiation of PC12 cells and contributed
to the analysis of the associated data. Brian B. Rudkin conceived and coordinated the study, contributed to the
analysis of the data, preparation of the figures and table and oversaw the writing of the manuscript along with
Chonggang Yuan. All authors reviewed the results and approved the final version of the manuscript.
Conflicts of Interest: The authors declare no conflict of interest.
Int. J. Mol. Sci. 2017, 18, 693 20 of 25

References
1. Simons, K.; Toomre, D. Lipid rafts and signal transduction. Nat. Rev. 2000, 1, 31–39. [CrossRef] [PubMed]
2. de Laurentiis, A.; Donovan, L.; Arcaro, A. Lipid rafts and caveolae in signaling by growth factor receptors.
Open Biochem. J. 2007, 1, 12–32. [CrossRef] [PubMed]
3. Parton, R.G.; Simons, K. The multiple faces of caveolae. Nat. Rev. 2007, 8, 185–194. [CrossRef] [PubMed]
4. Tang, Z.; Scherer, P.E.; Okamoto, T.; Song, K.; Chu, C.; Kohtz, D.S.; Nishimoto, I.; Lodish, H.F.; Lisanti, M.P.
Molecular cloning of caveolin-3, a novel member of the caveolin gene family expressed predominantly in
muscle. J. Biol. Chem. 1996, 271, 2255–2261. [PubMed]
5. Way, M.; Parton, R.G. M-caveolin, a muscle-specific caveolin-related protein. FEBS Lett. 1995, 376, 108–112.
[CrossRef]
6. Fra, A.M.; Williamson, E.; Simons, K.; Parton, R.G. De novo formation of caveolae in lymphocytes by
expression of VIP21-caveolin. Proc. Natl. Acad. Sci. USA 1995, 92, 8655–8659. [CrossRef] [PubMed]
7. Lipardi, C.; Mora, R.; Colomer, V.; Paladino, S.; Nitsch, L.; Rodriguez-Boulan, E.; Zurzolo, C.
Caveolin transfection results in caveolae formation but not apical sorting of glycosylphosphatidylinositol
(GPI)-anchored proteins in epithelial cells. J. Cell Biol. 1998, 140, 617–626. [CrossRef] [PubMed]
8. Sowa, G.; Xie, L.; Xu, L.; Sessa, W.C. Serine 23 and 36 phosphorylation of caveolin-2 is differentially regulated
by targeting to lipid raft/caveolae and in mitotic endothelial cells. Biochemistry 2008, 47, 101–111. [CrossRef]
[PubMed]
9. Okamoto, C.T. Endocytosis and transcytosis. Adv. Drug Deliv. Rev. 1998, 29, 215–228. [CrossRef]
10. Stary, C.; Tsutsumi, Y.; Patel, P.; Head, B.; Patel, H.; Roth, D. Caveolins: Targeting pro-survival signaling in
the heart and brain. Front. Physiol. 2012, 3, 393. [CrossRef] [PubMed]
11. Peiro, S.; Comella, J.X.; Enrich, C.; Martin-Zanca, D.; Rocamora, N. PC12 cells have caveolae that contain TrkA.
Caveolae-disrupting drugs inhibit nerve growth factor-induced, but not epidermal growth factor-induced,
MAPK phosphorylation. J. Biol. Chem. 2000, 275, 37846–37852. [CrossRef] [PubMed]
12. Huang, C.S.; Zhou, J.; Feng, A.K.; Lynch, C.C.; Klumperman, J.; DeArmond, S.J.; Mobley, W.C. Nerve growth
factor signaling in caveolae-like domains at the plasma membrane. J. Biol. Chem. 1999, 274, 36707–36714.
[CrossRef] [PubMed]
13. Hibbert, A.P.; Kramer, B.M.; Miller, F.D.; Kaplan, D.R. The localization, trafficking and retrograde transport
of BDNF bound to p75NTR in sympathetic neurons. Mol. Cell. Neurosci. 2006, 32, 387–402. [CrossRef]
[PubMed]
14. Bilderback, T.R.; Gazula, V.R.; Lisanti, M.P.; Dobrowsky, R.T. Caveolin interacts with Trk A and p75(NTR)
and regulates neurotrophin signaling pathways. J. Biol. Chem. 1999, 274, 257–263. [CrossRef] [PubMed]
15. Schmitz, M.; Kloppner, S.; Klopfleisch, S.; Mobius, W.; Schwartz, P.; Zerr, I.; Althaus, H.H. Mutual effects
of caveolin and nerve growth factor signaling in pig oligodendrocytes. J. Neurosci. Res. 2010, 88, 572–588.
[CrossRef] [PubMed]
16. Galbiati, F.; Volonte, D.; Gil, O.; Zanazzi, G.; Salzer, J.L.; Sargiacomo, M.; Scherer, P.E.; Engelman, J.A.;
Schlegel, A.; Parenti, M.; et al. Expression of caveolin-1 and -2 in differentiating PC12 cells and dorsal root
ganglion neurons: Caveolin-2 is up-regulated in response to cell injury. Proc. Natl. Acad. Sci. USA 1998, 95,
10257–10262. [CrossRef] [PubMed]
17. Wu, C.; Butz, S.; Ying, Y.; Anderson, R.G. Tyrosine kinase receptors concentrated in caveolae-like domains
from neuronal plasma membrane. J. Biol. Chem. 1997, 272, 3554–3559. [CrossRef] [PubMed]
18. Stern, C.M.; Mermelstein, P.G. Caveolin regulation of neuronal intracellular signaling. Cell. Mol. Life Sci.
2010, 67, 3785–3795. [CrossRef] [PubMed]
19. Gaudreault, S.; Dea, D.; Poirier, J. Increased caveolin-1 expression in Alzheimer’s disease brain.
Neurobiol. Aging 2004, 25, 753–759. [CrossRef] [PubMed]
20. Head, B.; Peart, J.; Panneerselvam, M.; Yokoyama, T.; Pearn, M.; Niesman, I.; Bonds, J.; Schilling, J.;
Miyanohara, A.; Headrick, J.; et al. Loss of caveolin-1 accelerates neurodegeneration and aging. PLoS ONE
2010, 5, 12. [CrossRef] [PubMed]
21. Galbiati, F.; Volonte, D.; Engelman, J.A.; Watanabe, G.; Burk, R.; Pestell, R.G.; Lisanti, M.P. Targeted
downregulation of caveolin-1 is sufficient to drive cell transformation and hyperactivate the p42/44 MAP
kinase cascade. EMBO J. 1998, 17, 6633–6648. [CrossRef] [PubMed]
Int. J. Mol. Sci. 2017, 18, 693 21 of 25

22. Burgermeister, E.; Liscovitch, M.; Rocken, C.; Schmid, R.M.; Ebert, M.P. Caveats of caveolin-1 in cancer
progression. Cancer Lett. 2008, 268, 187–201. [CrossRef] [PubMed]
23. Sainz-Jaspeado, M.; Martin-Liberal, J.; Lagares-Tena, L.; Mateo-Lozano, S.; Garcia del Muro, X.; Tirado, O.
Caveolin-1 in sarcomas: Friend or foe? Oncotarget 2011, 2, 305–312. [CrossRef] [PubMed]
24. Williams, T.M.; Lisanti, M.P. Caveolin-1 in oncogenic transformation, cancer, and metastasis. Am. J. Physiol.
Cell Physiol. 2005, 288, C494–C506. [CrossRef] [PubMed]
25. Sotgia, F.; Martinez-Outschoorn, U.E.; Pavlides, S.; Howell, A.; Pestell, R.G.; Lisanti, M.P. Understanding the
Warburg effect and the prognostic value of stromal caveolin-1 as a marker of a lethal tumor microenvironment.
Breast Cancer Res. 2011, 13, 213. [CrossRef] [PubMed]
26. Tahir, S.A.; Yang, G.; Goltsov, A.A.; Watanabe, M.; Tabata, K.; Addai, J.; el Fattah, M.A.; Kadmon, D.;
Thompson, T.C. Tumor cell-secreted caveolin-1 has proangiogenic activities in prostate cancer. Cancer Res.
2008, 68, 731–739. [CrossRef] [PubMed]
27. Patani, N.; Martin, L.A.; Reis-Filho, J.S.; Dowsett, M. The role of caveolin-1 in human breast cancer.
Breast Cancer Res. Treat. 2012, 131, 1–15. [CrossRef] [PubMed]
28. Logozzi, M.; De Milito, A.; Lugini, L.; Borghi, M.; Calabro, L.; Spada, M.; Perdicchio, M.; Marino, M.L.;
Federici, C.; Iessi, E.; et al. High levels of exosomes expressing CD63 and caveolin-1 in plasma of melanoma
patients. PLoS ONE 2009, 4, e5219. [CrossRef] [PubMed]
29. Kucharzewska, P.; Christianson, H.C.; Welch, J.E.; Svensson, K.J.; Fredlund, E.; Ringner, M.; Morgelin, M.;
Bourseau-Guilmain, E.; Bengzon, J.; Belting, M. Exosomes reflect the hypoxic status of glioma cells and
mediate hypoxia-dependent activation of vascular cells during tumor development. Proc. Natl. Acad.
Sci. USA 2013, 110, 7312–7317. [CrossRef] [PubMed]
30. Svensson, K.J.; Christianson, H.C.; Wittrup, A.; Bourseau-Guilmain, E.; Lindqvist, E.; Svensson, L.M.;
Morgelin, M.; Belting, M. Exosome uptake depends on ERK1/2-heat shock protein 27 signaling and lipid
Raft-mediated endocytosis negatively regulated by caveolin-1. J. Biol. Chem. 2013, 288, 17713–17724.
[CrossRef] [PubMed]
31. Syn, N.; Wang, L.; Sethi, G.; Thiery, J.P.; Goh, B.C. Exosome-Mediated Metastasis: From
Epithelial-Mesenchymal Transition to Escape from Immunosurveillance. Trends Pharmacol. Sci. 2016,
37, 606–617. [CrossRef] [PubMed]
32. Ernsberger, U. Role of neurotrophin signalling in the differentiation of neurons from dorsal root ganglia and
sympathetic ganglia. Cell Tissue Res. 2009, 336, 349–384. [CrossRef] [PubMed]
33. Ben-Zvi, A.; Ben-Gigi, L.; Yagil, Z.; Lerman, O.; Behar, O. Semaphorin3A regulates axon growth
independently of growth cone repulsion via modulation of TrkA signaling. Cell. Signal. 2008, 20, 467–479.
[CrossRef] [PubMed]
34. Lentz, S.I.; Knudson, C.M.; Korsmeyer, S.J.; Snider, W.D. Neurotrophins support the development of diverse
sensory axon morphologies. J. Neurosci. 1999, 19, 1038–1048. [PubMed]
35. Liu, R.Y.; Schmid, R.S.; Snider, W.D.; Maness, P.F. NGF enhances sensory axon growth induced by laminin
but not by the L1 cell adhesion molecule. Mol. Cell. Neurosci. 2002, 20, 2–12. [CrossRef] [PubMed]
36. White, F.A.; Silos-Santiago, I.; Molliver, D.C.; Nishimura, M.; Phillips, H.; Barbacid, M.; Snider, W.D.
Synchronous onset of NGF and TrkA survival dependence in developing dorsal root ganglia. J. Neurosci.
1996, 16, 4662–4672. [PubMed]
37. Yu, L.; Reynaud, F.; Falk, J.; Spencer, A.; Ding, Y.D.; Baumle, V.; Lu, R.; Castellani, V.; Yuan, C.; Rudkin, B.B.
Highly efficient method for gene delivery into mouse dorsal root ganglia neurons. Front. Mol. Neurosci. 2015,
8, 2. [CrossRef] [PubMed]
38. Tagawa, A.; Mezzacasa, A.; Hayer, A.; Longatti, A.; Pelkmans, L.; Helenius, A. Assembly and trafficking of
caveolar domains in the cell: Caveolae as stable, cargo-triggered, vesicular transporters. J. Cell Biol. 2005,
170, 769–779. [CrossRef] [PubMed]
39. Greene, L.A.; Tischler, A.S. Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma
cells which respond to nerve growth factor. Proc. Natl. Acad. Sci. USA 1976, 73, 2424–2428. [CrossRef]
[PubMed]
40. Rudkin, B.B.; Lazarovici, P.; Levi, B.Z.; Abe, Y.; Fujita, K.; Guroff, G. Cell cycle-specific action of nerve growth
factor in PC12 cells: Differentiation without proliferation. EMBO J. 1989, 8, 3319–3325. [PubMed]
41. Yan, G.Z.; Ziff, E.B. NGF regulates the PC12 cell cycle machinery through specific inhibition of the Cdk
kinases and induction of cyclin D1. J. Neurosci. 1995, 15, 6200–6212. [PubMed]
Int. J. Mol. Sci. 2017, 18, 693 22 of 25

42. Van Grunsven, L.A.; Billon, N.; Savatier, P.; Thomas, A.; Urdiales, J.L.; Rudkin, B.B. Effect of nerve growth
factor on the expression of cell cycle regulatory proteins in PC12 cells: Dissection of the neurotrophic
response from the anti-mitogenic response. Oncogene 1996, 12, 1347–1356. [PubMed]
43. Datto, M.B.; Li, Y.; Panus, J.F.; Howe, D.J.; Xiong, Y.; Wang, X.F. Transforming growth factor beta induces the
cyclin-dependent kinase inhibitor p21 through a p53-independent mechanism. Proc. Natl. Acad. Sci. USA
1995, 92, 5545–5549. [CrossRef] [PubMed]
44. Billon, N.; Carlisi, D.; Datto, M.B.; van Grunsven, L.A.; Watt, A.; Wang, X.F.; Rudkin, B.B. Cooperation
of Sp1 and p300 in the induction of the CDK inhibitor p21WAF1/CIP1 during NGF-mediated neuronal
differentiation. Oncogene 1999, 18, 2872–2882. [CrossRef] [PubMed]
45. Billon, N.; van Grunsven, L.A.; Rudkin, B.B. The CDK inhibitor p21WAF1/Cip1 is induced through a
p300-dependent mechanism during NGF-mediated neuronal differentiation of PC12 cells. Oncogene 1996, 13,
2047–2054. [PubMed]
46. Perrone, L.; Paladino, S.; Mazzone, M.; Nitsch, L.; Gulisano, M.; Zurzolo, C. Functional interaction between
p75NTR and TrkA: The endocytic trafficking of p75NTR is driven by TrkA and regulates TrkA-mediated
signalling. Biochem. J. 2005, 385, 233–241. [CrossRef] [PubMed]
47. Urdiales, J.L.; Becker, E.; Andrieu, M.; Thomas, A.; Jullien, J.; van Grunsven, L.A.; Menut, S.; Evan, G.I.;
Martin-Zanca, D.; Rudkin, B.B. Cell cycle phase-specific surface expression of nerve growth factor receptors
TrkA and p75(NTR). J. Neurosci. 1998, 18, 6767–6775. [PubMed]
48. Xu, X.H.; Zhang, Z.G.; Wu, F.M. [Improving effect of NGF by intrahippocampal injection on aging-related
memory impairment]. Yao Xue Xue Bao 2000, 35, 729–732. [PubMed]
49. 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.;
Mobley, W.C. Endocytosis of activated TrkA: Evidence that nerve growth factor induces formation of
signaling endosomes. J. Neurosci. 1996, 16, 7950–7964. [PubMed]
50. Grimes, M.L.; Beattie, E.; Mobley, W.C. A signaling organelle containing the nerve growth factor-activated
receptor tyrosine kinase, TrkA. Proc. Natl. Acad. Sci. USA 1997, 94, 9909–9914. [CrossRef] [PubMed]
51. MacInnis, B.L.; Senger, D.L.; Campenot, R.B. Spatial requirements for TrkA kinase activity in the support
of neuronal survival and axon growth in rat sympathetic neurons. Neuropharmacology 2003, 45, 995–1010.
[CrossRef]
52. Senger, D.L.; Campenot, R.B. Rapid retrograde tyrosine phosphorylation of trkA and other proteins in rat
sympathetic neurons in compartmented cultures. J. Cell. Biol. 1997, 138, 411–421. [CrossRef] [PubMed]
53. Harrington, A.W.; St Hillaire, C.; Zweifel, L.S.; Glebova, N.O.; Philippidou, P.; Halegoua, S.; Ginty, D.D.
Recruitment of actin modifiers to TrkA endosomes governs retrograde NGF signaling and survival. Cell
2011, 146, 421–434. [CrossRef] [PubMed]
54. Bhattacharyya, A.; Watson, F.L.; Bradlee, T.A.; Pomeroy, S.L.; Stiles, C.D.; Segal, R.A. Trk receptors function
as rapid retrograde signal carriers in the adult nervous system. J. Neurosci. 1997, 17, 7007–7016. [PubMed]
55. Jullien, J.; Guili, V.; Derrington, E.A.; Darlix, J.L.; Reichardt, L.F.; Rudkin, B.B. Trafficking of TrkA-green
fluorescent protein chimerae during nerve growth factor-induced differentiation. J. Biol. Chem. 2003, 278,
8706–8716. [CrossRef] [PubMed]
56. Clary, D.O.; Weskamp, G.; Austin, L.R.; Reichardt, L.F. TrkA cross-linking mimics neuronal responses to
nerve growth factor. Mol. Biol. Cell 1994, 5, 549–563. [CrossRef] [PubMed]
57. Barker, P.A.; Shooter, E.M. Disruption of NGF binding to the low affinity neurotrophin receptor p75LNTR
reduces NGF binding to TrkA on PC12 cells. Neuron 1994, 13, 203–215. [CrossRef]
58. Jullien, J.; Guili, V.; Reichardt, L.F.; Rudkin, B.B. Molecular kinetics of nerve growth factor receptor trafficking
and activation. J. Biol. Chem. 2002, 277, 38700–38708. [CrossRef] [PubMed]
59. Loeb, D.M.; Maragos, J.; Martin-Zanca, D.; Chao, M.V.; Parada, L.F.; Greene, L.A. The trk proto-oncogene
rescues NGF responsiveness in mutant NGF-nonresponsive PC12 cell lines. Cell 1991, 66, 961–966. [CrossRef]
60. Marshall, C.J. Specificity of receptor tyrosine kinase signaling: Transient versus sustained extracellular
signal-regulated kinase activation. Cell 1995, 80, 179–185. [CrossRef]
61. Ginty, D.D.; Bonni, A.; Greenberg, M.E. Nerve growth factor activates a Ras-dependent protein kinase that
stimulates c-fos transcription via phosphorylation of CREB. Cell 1994, 77, 713–725. [CrossRef]
62. Xing, J.; Ginty, D.D.; Greenberg, M.E. Coupling of the RAS-MAPK pathway to gene activation by RSK2, a
growth factor-regulated CREB kinase. Science 1996, 273, 959–963. [CrossRef] [PubMed]
Int. J. Mol. Sci. 2017, 18, 693 23 of 25

63. Descamps, S.; Toillon, R.A.; Adriaenssens, E.; Pawlowski, V.; Cool, S.M.; Nurcombe, V.; Le Bourhis, X.;
Boilly, B.; Peyrat, J.P.; Hondermarck, H. Nerve growth factor stimulates proliferation and survival of human
breast cancer cells through two distinct signaling pathways. J. Biol. Chem. 2001, 276, 17864–17870. [CrossRef]
[PubMed]
64. Dolle, L.; Adriaenssens, E.; El Yazidi-Belkoura, I.; Le Bourhis, X.; Nurcombe, V.; Hondermarck, H. Nerve
growth factor receptors and signaling in breast cancer. Curr. Cancer Drug Target 2004, 4, 463–470. [CrossRef]
65. Li, T.; Sotgia, F.; Vuolo, M.A.; Li, M.; Yang, W.C.; Pestell, R.G.; Sparano, J.A.; Lisanti, M.P. Caveolin-1
mutations in human breast cancer: Functional association with estrogen receptor alpha-positive status.
Am. J. Pathol. 2006, 168, 1998–2013. [CrossRef] [PubMed]
66. Nico, B.; Mangieri, D.; Benagiano, V.; Crivellato, E.; Ribatti, D. Nerve growth factor as an angiogenic factor.
Microvasc. Res. 2008, 75, 135–141. [CrossRef] [PubMed]
67. Zhu, Z.; Kleeff, J.; Kayed, H.; Wang, L.; Korc, M.; Buchler, M.W.; Friess, H. Nerve growth factor and
enhancement of proliferation, invasion, and tumorigenicity of pancreatic cancer cells. Mol. Carcinog. 2002,
35, 138–147. [CrossRef] [PubMed]
68. Tahir, S.A.; Yang, G.; Ebara, S.; Timme, T.L.; Satoh, T.; Li, L.; Goltsov, A.; Ittmann, M.; Morrisett, J.D.;
Thompson, T.C. Secreted caveolin-1 stimulates cell survival/clonal growth and contributes to metastasis in
androgen-insensitive prostate cancer. Cancer Res. 2001, 61, 3882–3885. [PubMed]
69. Hayashi, K.; Matsuda, S.; Machida, K.; Yamamoto, T.; Fukuda, Y.; Nimura, Y.; Hayakawa, T.; Hamaguchi, M.
Invasion activating caveolin-1 mutation in human scirrhous breast cancers. Cancer Res. 2001, 61, 2361–2364.
[PubMed]
70. Schlegel, A.; Arvan, P.; Lisanti, M.P. Caveolin-1 binding to endoplasmic reticulum membranes and entry
into the regulated secretory pathway are regulated by serine phosphorylation. Protein sorting at the level of
the endoplasmic reticulum. J. Biol. Chem. 2001, 276, 4398–4408. [CrossRef] [PubMed]
71. Sonnino, S.; Prinetti, A. Membrane domains and the “Lipid Raft” concept. Curr. Med. Chem. 2013, 20, 4–21.
[PubMed]
72. Pike, L.J. Growth factor receptors, lipid rafts and caveolae: An evolving story. Biochim. Biophys. Acta 2005,
1746, 260–273. [CrossRef] [PubMed]
73. Zonta, B.; Minichiello, L. Synaptic membrane rafts: Traffic lights for local neurotrophin signaling?
Front. Synaptic. Neurosci. 2013, 5, 9. [CrossRef] [PubMed]
74. Paratcha, G.; Ibanez, C.F. Lipid rafts and the control of neurotrophic factor signaling in the nervous system:
Variations on a theme. Curr. Opin. Neurobiol. 2002, 12, 542–549. [CrossRef]
75. Irschick, R.; Trost, T.; Karp, G.; Hausott, B.; Auer, M.; Claus, P.; Klimaschewski, L. Sorting of the FGF receptor
1 in a human glioma cell line. Histochem. Cell. Biol. 2013, 139, 135–148. [CrossRef] [PubMed]
76. Boothe, T.; Lim, G.E.; Cen, H.; Skovsø, S.; Piske, M.; Li, S.N.; Nabi, I.R.; Gilon, P.; Johnson, J.D. Inter-domain
tagging implicates caveolin-1 in insulin receptor trafficking and Erk signaling bias in pancreatic beta-cells.
Mol. Metab. 2016, 5, 366–378. [CrossRef] [PubMed]
77. Limpert, A.S.; Karlo, J.C.; Landreth, G.E. Nerve growth factor stimulates the concentration of TrkA
within lipid rafts and extracellular signal-regulated kinase activation through c-Cbl-associated protein.
Mol. Cell. Biol. 2007, 27, 5686–5698. [CrossRef] [PubMed]
78. Pereira, D.B.; Chao, M.V. The tyrosine kinase Fyn determines the localization of TrkB receptors in lipid rafts.
J. Neurosci. 2007, 27, 4859–4869. [CrossRef] [PubMed]
79. Pryor, S.; McCaffrey, G.; Young, L.R.; Grimes, M.L. NGF causes TrkA to specifically attract microtubules to
lipid rafts. PLoS ONE 2012, 7, e35163.
80. Nishio, M.; Fukumoto, S.; Furukawa, K.; Ichimura, A.; Miyazaki, H.; Kusunoki, S.; Urano, T. Overexpressed
GM1 suppresses nerve growth factor (NGF) signals by modulating the intracellular localization of NGF
receptors and membrane fluidity in PC12 cells. J. Biol. Chem. 2004, 279, 33368–33378. [CrossRef] [PubMed]
81. Ingelmo-Torres, M.; Gaus, K.; Herms, A.; Gonzalez-Moreno, E.; Kassan, A.; Bosch, M.; Grewal, T.; Tebar, F.;
Enrich, C.; Pol, A. Triton X-100 promotes a cholesterol-dependent condensation of the plasma membrane.
Biochem. J. 2009, 420, 373–381. [CrossRef] [PubMed]
82. Owen, D.M.; Gaus, K. Imaging lipid domains in cell membranes: The advent of super-resolution fluorescence
microscopy. Front. Plant Sci. 2013, 4, 503. [CrossRef] [PubMed]
83. Ma, Y.; Hinde, E.; Gaus, K. Nanodomains in biological membranes. Essays Biochem. 2015, 57, 93–107.
[CrossRef] [PubMed]
Int. J. Mol. Sci. 2017, 18, 693 24 of 25

84. Kim, Y.N.; Bertics, P.J. The endocytosis-linked protein dynamin associates with caveolin-1 and is tyrosine
phosphorylated in response to the activation of a noninternalizing epidermal growth factor receptor mutant.
Endocrinology 2002, 143, 1726–1731. [CrossRef] [PubMed]
85. Pike, L.J. Rafts defined: A report on the Keystone Symposium on Lipid Rafts and Cell Function. J. Lipid. Res.
2006, 47, 1597–1598. [CrossRef] [PubMed]
86. Lajoie, P.; Kojic, L.D.; Nim, S.; Li, L.; Dennis, J.W.; Nabi, I.R. Caveolin-1 regulation of dynamin-dependent,
raft-mediated endocytosis of cholera toxin-B sub-unit occurs independently of caveolae. J. Cell. Mol. Med.
2009, 13, 3218–3225. [CrossRef] [PubMed]
87. Fielding, P.E.; Chau, P.; Liu, D.; Spencer, T.A.; Fielding, C.J. Mechanism of platelet-derived growth
factor-dependent caveolin-1 phosphorylation: Relationship to sterol binding and the role of serine-80.
Biochemistry 2004, 43, 2578–2586. [CrossRef] [PubMed]
88. Couet, J.; Sargiacomo, M.; Lisanti, M.P. Interaction of a receptor tyrosine kinase, EGF-R, with caveolins.
Caveolin binding negatively regulates tyrosine and serine/threonine kinase activities. J. Biol. Chem. 1997,
272, 30429–30438. [CrossRef] [PubMed]
89. Bilderback, T.R.; Grigsby, R.J.; Dobrowsky, R.T. Association of p75(NTR) with caveolin and localization
of neurotrophin-induced sphingomyelin hydrolysis to caveolae. J. Biol. Chem. 1997, 272, 10922–10927.
[PubMed]
90. Kimpinski, K.; Jelinski, S.; Mearow, K. The anti-p75 antibody, MC192, and brain-derived neurotrophic factor
inhibit nerve growth factor-dependent neurite growth from adult sensory neurons. Neuroscience 1999, 93,
253–263. [CrossRef]
91. Lahtinen, U.; Honsho, M.; Parton, R.G.; Simons, K.; Verkade, P. Involvement of caveolin-2 in caveolar
biogenesis in MDCK cells. FEBS Lett. 2003, 538, 85–88. [CrossRef]
92. Razani, B.; Wang, X.B.; Engelman, J.A.; Battista, M.; Lagaud, G.; Zhang, X.L.; Kneitz, B.; Hou, H., Jr.;
Christ, G.J.; Edelmann, W.; et al. Caveolin-2-deficient mice show evidence of severe pulmonary dysfunction
without disruption of caveolae. Mol. Cell. Biol. 2002, 22, 2329–2344. [CrossRef] [PubMed]
93. Scherer, P.E.; Lewis, R.Y.; Volonte, D.; Engelman, J.A.; Galbiati, F.; Couet, J.; Kohtz, D.S.; van Donselaar, E.;
Peters, P.; Lisanti, M.P. Cell-type and tissue-specific expression of caveolin-2. Caveolins 1 and 2 co-localize
and form a stable hetero-oligomeric complex in vivo. J. Biol. Chem. 1997, 272, 29337–29346. [CrossRef]
[PubMed]
94. Chan, Y.G.; Cardwell, M.M.; Hermanas, T.M.; Uchiyama, T.; Martinez, J.J. Rickettsial outer-membrane protein
B (rOmpB) mediates bacterial invasion through Ku70 in an actin, c-Cbl, clathrin and caveolin 2-dependent
manner. Cell. Microbiol. 2009, 11, 629–644. [CrossRef] [PubMed]
95. Parker, S.; Walker, D.S.; Ly, S.; Baylis, H.A. Caveolin-2 is required for apical lipid trafficking and suppresses
basolateral recycling defects in the intestine of Caenorhabditis elegans. Mol. Biol. Cell 2009, 20, 1763–1771.
[CrossRef] [PubMed]
96. Fong, A.; Garcia, E.; Gwynn, L.; Lisanti, M.P.; Fazzari, M.J.; Li, M. Expression of caveolin-1 and caveolin-2
in urothelial carcinoma of the urinary bladder correlates with tumor grade and squamous differentiation.
Am. J. Clin. Pathol. 2003, 120, 93–100. [CrossRef] [PubMed]
97. Hu, Y.C.; Lam, K.Y.; Law, S.; Wong, J.; Srivastava, G. Profiling of differentially expressed cancer-related
genes in esophageal squamous cell carcinoma (ESCC) using human cancer cDNA arrays: Overexpression of
oncogene MET correlates with tumor differentiation in ESCC. Clin. Cancer Res. 2001, 7, 3519–3525. [PubMed]
98. Kwon, H.; Pak, Y. Prolonged tyrosine kinase activation of insulin receptor by pY27-caveolin-2.
Biochem. Biophys. Res. Commun. 2009. [CrossRef] [PubMed]
99. Mora, R.; Bonilha, V.L.; Marmorstein, A.; Scherer, P.E.; Brown, D.; Lisanti, M.P.; Rodriguez-Boulan, E.
Caveolin-2 localizes to the golgi complex but redistributes to plasma membrane, caveolae, and rafts when
co-expressed with caveolin-1. J. Biol. Chem. 1999, 274, 25708–25717. [CrossRef] [PubMed]
100. Tisi, M.A.; Xie, Y.; Yeo, T.T.; Longo, F.M. Downregulation of LAR tyrosine phosphatase prevents apoptosis
and augments NGF-induced neurite outgrowth. J. Neurobiol. 2000, 42, 477–486. [CrossRef]
101. Meijering, E.; Jacob, M.; Sarria, J.C.; Steiner, P.; Hirling, H.; Unser, M. Design and validation of a tool for
neurite tracing and analysis in fluorescence microscopy images. Cytom. Part A 2004, 58, 167–176. [CrossRef]
[PubMed]
Int. J. Mol. Sci. 2017, 18, 693 25 of 25

102. Yan, G.Z.; Ziff, E.B. Nerve growth factor induces transcription of the p21 WAF1/CIP1 and cyclin D1 genes in
PC12 cells by activating the Sp1 transcription factor. J. Neurosci. 1997, 17, 6122–6132. [PubMed]
103. Song, K.S.; Li, S.; Okamoto, T.; Quilliam, L.A.; Sargiacomo, M.; Lisanti, M.P. Co-purification and direct
interaction of Ras with caveolin, an integral membrane protein of caveolae microdomains. Detergent-free
purification of caveolae microdomains. J. Biol. Chem. 1996, 271, 9690–9697. [PubMed]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).