Epidermal growth factor receptor endocytic traffic

perturbation by phosphatidate phosphohydrolase

inhibition: new strategy against cancer
Ronan Shaughnessy1,2, Claudio Retamal1,2, Claudia Oyanadel1,2, Andres Norambuena1,2,
 pez , Marcela Bravo-Zehnder , Fabian J. Montecino , Claudia Metz1,2,
Alejandro Lo 1,2 1,2 1,2

Andrea Soza1,2 and Alfonso Gonzalez1,2

lica de Chile, Santiago, Chile
1 Departamento de Inmunologıa Clınica y Reumatologıa, Facultad de Medicina, Pontificia Universidad Cato
2 Centro de Envejecimiento y Regeneracio n, Departamento de Biologıa Celular y Molecular, Facultad de Ciencias Biolo
gicas, Pontificia
lica de Chile, Santiago, Chile
Universidad Cato

Keywords Epidermal growth factor receptor (EGFR) exaggerated (oncogenic) func-

apoptosis; cancer; EGFR; endocytosis;
phosphatidic acid
Correspondence
A. González, Departamento de Inmunologı́a
Clı́nica y Reumatologı́a, Facultad de
Medicina, Marcoleta 367, Pontificia
Universidad Católica de Chile, 8330025
Santiago, Chile
Fax: +56 2 2229995
Tel: +56 2 26862713
E-mail: agonzara@med.puc.cl
(Received 12 July 2013, revised 2 February
2014, accepted 26 February 2014)
doi:10.1111/febs.12770
tion is currently targeted in cancer treatment with drugs that block recep-
tor ligand binding or tyrosine kinase activity. Because endocytic trafficking
is a crucial regulator of EGFR function, its pharmacological perturbation
might provide a new anti-tumoral strategy. Inhibition of phosphatidic acid
(PA) phosphohydrolase (PAP) activity has been shown to trigger PA sig-
naling towards type 4 phosphodiesterase (PDE4) activation and protein
kinase A inhibition, leading to internalization of empty/inactive EGFR.
Here, we used propranolol, its L- and D- isomers and desipramine as PAP
inhibitors to further explore the effects of PAP inhibition on EGFR endo-
cytic trafficking and its consequences on EGFR-dependent cancer cell line
models. PAP inhibition not only made EGFR inaccessible to stimuli but
also prolonged the signaling lifetime of ligand-activated EGFR in recycling
endosomes. Strikingly, such endocytic perturbations applied in acute/inter-
mittent PAP inhibitor treatments selectively impaired cell proliferation/
viability sustained by an exaggerated EGFR function. Phospholipase D
inhibition with FIPI (5-fluoro-2-indolyl des-chlorohalopemide) and PDE4
inhibition with rolipram abrogated both the anti-tumoral and endocytic
effects of PAP inhibition. Prolonged treatments with a low concentration
of PAP inhibitors, although without detectable endocytic effects, still coun-
teracted cell proliferation, induced apoptosis and decreased anchorage-
independent growth of cells bearing EGFR oncogenic influences. Overall,
our results show that PAP inhibitors can counteract EGFR oncogenic
traits, including receptor overexpression or activating mutations resistant
to current tyrosine kinase inhibitors, perturbing EGFR endocytic traffick-
ing and perhaps other as yet unknown processes, depending on treatment
conditions. This puts PAP activity forward as a new suitable target against
EGFR-driven malignancy.

Abbreviations
DAG, diacylglycerol; EEA1, early endosome antigen 1; EGF, epidermal growth factor; EGFR, EGF receptor; GFP, green fluorescent protein;
LPP, lipid phosphate phosphatase; PA, phosphatidic acid; PAP, phosphatidic acid phosphohydrolase; PARP, poly ADP ribose polymerase;
PDE4, type 4 phosphodiesterases; PKA, protein kinase A; PLD, phospholipase D; TfR, transferrin receptor.

2172 FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS

R. Shaughnessy et al.
Introduction
The epidermal growth factor receptor (EGFR) belongs
to the family of ErbB tyrosine kinase receptors that
regulate processes of cell proliferation, differentiation,
migration and survival, playing crucial roles in devel-
opment, tissue repair and cancer [1,2]. An exaggerated
(oncogenic) activity due to EGFR overexpression or
activating mutations is frequently found in tumor cells,
contributing to their malignancy. A major challenge
for targeted anti-cancer treatments is to interfere with
such oncogenic EGFR function, which gives an oppor-
tunity to attack cancerous cells with minimal damage
to normal cells [2,3]. Current drugs have so far been
directly targeted to the EGFR molecule, either to
block ligand binding (antibodies) or to inhibit its tyro-
sine kinase activity (small molecules), both demon-
strating therapeutic effectiveness against certain
EGFR-dependent cancers [2]. Limitations such as low
response rates and development of resistance prompt a
search for new strategies to counteract the malignant
influence of the EGFR [4]. An interesting, yet little
explored, alternative is to perturb EGFR endocytic
trafficking.
Endocytic trafficking has gained preponderance in
EGFR function control [5], offering multiple opportu-
nities to modify the intensity, location and duration of
receptor signaling [5–9]. Upon ligand binding, EGFRs
undergo dimerization, tyrosine kinase activation, trans-
phosphorylation and ubiquitylation, altogether con-
tributing to define the endocytic trafficking of active
EGFRs, which adds spatiotemporal dynamics to sig-
naling [5,10–14]. Depending on ligand concentration
and threshold-controlled ubiquitylation [11], activated
EGFRs can follow a clathrin-dependent/recycling
route, which prolongs signaling activity, or a clathrin-
independent/ubiquitination-dependent downregulation
pathway, which avoids excessive signaling [11,15].
Downregulation involves ESCRT-mediated sorting
into intraluminal vesicles of multivesicular bodies that
then fuse with lysosomes [16]. Thus, activated EGFR
can remain signaling-competent for variable periods of
time and at different cellular locations before degrada-
tion, specifying different response outcomes [15].
Cancerous cells seem to exploit the endocytic traf-
ficking machinery to increase EGFR oncogenic activity
[17–19], and thus in principle might be more sensitive
than normal cells to endocytic perturbations. A recent
study [20] suggests that signaling imbalances with dele-
terious actions on EGFR-dependent cancer cells can
be elicited by delaying the traffic of ligand-activated
EGFRs into the lysosomal degradation pathway. The
mechanisms that regulate EGFR signaling, as well as
FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS
Targeting endocytosis blocks oncogenic EGFR
other signaling receptors, are intertwined with the
mechanisms that regulate endocytic trafficking [6–
9,21], involving a variety of kinases from diverse
signaling pathways [22]. Therefore, exploring the
potential anti-tumoral actions of drugs previously
reported to alter EGFR endocytic traffic, acting upon
protein kinase A (PKA) [23], seems quite attractive.
Inhibition of basal PKA activity not only delays
sorting of ligand-activated EGFR to the degradation
pathway but also induces internalization of empty/
inactive EGFRs making them inaccessible to stimula-
tion [24]. These two endocytic perturbations might be
expected to exert deleterious effects on EGFR-depen-
dent malignant cells. PKA and its regulatory systems
are intensively studied in the search for drug design
strategies, which include targeting the synthesis and
degradation of cAMP [25]. Indeed, the most known
signaling pathway towards decreasing PKA activity is
the activation of Gai proteins by certain G-protein-
coupled receptors, but an effective intervention
through this system would depend on which of these
receptors are expressed [26]. A more suitable approach
would be to target the ubiquitous signaling pathway of
phosphatidic acid (PA), which for a long time has
been described to decrease PKA activity [27]. This PA/
PKA pathway can be easily triggered by cationic
amphiphilic compounds currently used in clinical prac-
tice, such as propranolol and desipramine [28], and
has recently arisen as a new control system of EGFR
endocytosis [23].
PA is a phospholipid present in small amounts in bio-
logical membranes where its levels can be regulated by
several enzymes [29]. Besides serving as a lipid synthesis
precursor, PA is also involved in signal transduction,
pH sensing and membrane bending, which are function-
ally projected in processes such as cell proliferation, sur-
vival, apoptosis and migration [29], as well as in
membrane trafficking along the exocytic and endocytic
routes [30]. PA has been involved in the endocytosis of
ligand-activated EGFR [31,32], as well as in PKA-regu-
lated endocytosis of empty/inactive EGFR [23]. Phos-
pholipase D (PLD) is a main generator of PA in
response to different extracellular stimuli [33], while
phosphatidic acid phosphohydrolases (PAPs) are
important PA downregulators, producing diacylglycerol
(DAG) [34]. PA signaling functions are mediated by
producing other signaling lipids, such as DAG, lyso-
phosphatidic acid and phospoinositides [29,33], and by
recruiting and activating specific signaling proteins bear-
ing PA binding domains [29,33]. An interesting example
of such PA binding proteins are type 4 phosphodiester-
ases (PDE4) [27], which are the main regulators of
cAMP levels and compartmentalized cAMP signaling
2173
Targeting endocytosis blocks oncogenic EGFR
[35,36]. Propranolol and desipramine used as PAP
inhibitors transiently increase PA levels leading to
PDE4 activation and the subsequent reduction of
cAMP levels and PKA activity [27]. Recent evidence
shows that such a PA/PDE4/PKA signaling pathway
induces endocytic removal of empty/inactive EGFRs
from the cell surface via both clathrin-dependent and
clathrin-independent routes, resulting in their reversible
accumulation in recycling endosomes [23]. The obvious
remaining question is whether PAP inhibitors might eli-
cit endocytic perturbations with selective deleterious
consequences upon EGFR-dependent tumoral cells.
Results
Various PAP inhibitors induce EGFR
internalization in the absence of ligand
Propranolol used as a PAP inhibitor induces internali-
zation of empty/inactive EGFR, making it inaccessible
to external stimuli [23]. This suggests the possibility of
interfering with EGFR oncogenic function using PAP
inhibitors. However, specific PAP inhibitors are not
available and therefore the studies are based on cat-
ionic amphiphilic compounds, such as propranolol and
desipramine, currently used in clinical practice for
other purposes [28,34,37]. Propranolol is a racemic
mixture of D- and L-propranolol isomers [38,39], which
are equivalent as inhibitors of PAP activity, as well as
inducers of the PA/PDE4/PKA pathway [27,37]. Simi-
larly, desipramine inhibits PAP activity but to a
greater extent than propranolol [37] and also triggers
the PA/PDE4/PKA pathway [27]. Therefore, using our
previously characterized ‘in-house’ HeLa cells that
express ~ 300 000 EGFR
cell 1 [23,24], we first ana-
lyzed the potency of L- and D-propranolol isomers and
desipramine upon EGFR internalization in the absence
of ligand.
HeLa cells were incubated with increasing concen-
trations of each drug for 30 min at 37 °C and then
radioligand binding was performed at 4 °C to assess
EGFRs remaining at the cell surface (Fig. 1A). D- and
L-propranolol were equivalent as inducers of EGFR
internalization, both displaying an EC50 of 75–100 lM,
similar to that described for the propranolol racemic
mixture [23]. Desipramine was more potent (EC50 of
20 lM) than propranolol at inducing EGFR internali-
zation, thus correlating with its higher potency as a
PAP inhibitor [37]. Such EC50 concentrations are 10-
fold lower than those of propranolol used to com-
pletely inhibit PAP activity in cells [40,41]. Additive
effects were obtained with a mixture of 75 lM pro-
pranolol and 20 lM desipramine (Fig. 1A).
2174
R. Shaughnessy et al.
As previously reported for propranolol and PKA
inhibitors [23,24], indirect immunofluorescence showed
that all drugs induce EGFR redistribution from the
cell surface towards perinuclear endocytic compart-
ments (Fig. 1B). Therefore, propranolol and its iso-
mers can be used indistinctly in the endocytic
experiments. In cell proliferation assays we used D-pro-
pranolol, which has 60–100-fold less b-adrenergic
blocker activity than L-propranolol [38,39].
PAP inhibition delays degradation and increases
the pool of ligand-activated EGFR in endosomes
To study the effect of PAP inhibition on the degrada-
tion kinetics and endocytic trafficking of ligand-acti-
vated EGFR, we used our ‘in-house’ HeLa cells as well
as HeLa cells from the ATCC (HeLa-wt) permanently
transfected to overexpress EGFR-GFP (HeLa-EGFR)
at almost the same level as our ‘in-house’ cells. HeLa-
wt cells from the ATCC have been reported to express
around 35 000–65 000 EGFR
cell 1 [15] and therefore
are less convenient for assessing EGFR distribution.
Treatment of ‘in-house’ HeLa cells with epidermal
growth factor (EGF) (50 ng
mL 1) induced a gradual
decrease in EGFR mass, with a half-life of about
100 min, reaching 70% degradation in 3 h. The pres-
ence of propranolol stabilized the initial slope of deg-
radation to about 30% within the first hour and was
then maintained for the rest of the experiment (4 h)
(Fig. 2A). Tyrosine phosphorylation reflecting EGFR
signaling activity showed a prolonged lifetime under
propranolol treatment (Fig. 2B). After 1 h of propran-
olol treatment we could detect an increment in the
pool of EGFR colocalizing with endosomes containing
either early endosome antigen 1 (EEA1) or transferrin
receptor (TfR) (Fig. 2C).
Taken together with the immunoblot analysis, these
results indicate that PAP inhibition by propranolol not
only induces EGFR internalization of empty/inactive
receptors, as previously shown, but also delays sorting
of ligand-activated EGFR towards the lysosomal deg-
radation route. The increased pool of EGFRs in early/
recycling endosomes, mostly in perinuclear recycling
endosomes, probably include ligand-activated recep-
tors. This is shown later in the HeLa-EGFR cells
(Fig. 4, later).
PAP inhibition has anti-proliferative/viability
effects that require EGFR endocytosis
Recent studies have shown that induced increases in
the pool of active EGFRs in endosomal compart-
ments can be deleterious to EGFR-dependent tumoral
FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS
R. Shaughnessy et al. Targeting endocytosis blocks oncogenic EGFR

Fig. 1. PAP inhibitors induce EGFR

internalization in the absence of ligand. (A)
125
EGF binding was performed at 4 °C in
HeLa cells treated for 30 min with B
increasing concentrations of D-propranolol,
L-propranolol and desipramine (DMI). The

decrease in binding activity, which reflects

EGFR endocytosis [23], shows similar
EC50 (75–100 lM) for D- and L-propranolol
and lower EC50 (20–25 lM) for
desipramine. A combination of 75 lM of
racemic propranolol and 20 lM
desipramine show additive effects on
EGFR inaccessibility (internalization). (B)
Immunofluorescent pattern showing
internalized EGFR. HeLa cells were treated
with 100 lM racemic propranolol,
D-propranolol, L-propranolol or 20 lM
desipramine for 30 min.
Immunofluorescence with the monoclonal
HB8506 anti-EGFR shows similar
perinuclear location of EGFR with all
drugs.

cells [20]. To study the functional consequences of the
endocytic perturbations induced by PAP inhibition we
compared HeLa-wt with permanently transfected
HeLa-EGFR that mimic recently acquired overexpres-
sion of EGFR as an oncogenic trait. HeLa-EGFR
cells show higher rates of proliferation than their
parental HeLa-wt (ATCC) counterparts (Fig. 3A) and
express similar levels of EGFR to our ‘in-house’
HeLa cells (Fig. 3B). D-propranolol at the EC50 for
EGFR internalization increased the levels of PA by
~ 60% in 20 min (Fig. 3C). Strikingly, these HeLa-
EGFR cells showed higher sensitivity to acute and
intermittent treatments with D-propranolol than
HeLa-wt and our ‘in-house’ HeLa cells (Fig. 3D). The
treatment consisted of incubations with 75 lM D-pro-
pranolol for just 1 h, which, as shown, induces EGFR
trafficking perturbations, repeated twice a day with
intervals of 6 h without drugs for 4 days. Under
these conditions, proliferation/viability significantly
decreased in HeLa-EGFR cells, while HeLa cells
showed a slight decrease without reaching statistical
significance (Fig. 3D). This treatment was also ineffec-
tive upon non-tumoral epithelial MDCK cells (not
shown), which express very low levels of EGFR and
represent conditions close to normal epithelial cells
[42]. Inhibition of PLD activity with FIPI (5-fluoro-
2-indolyl des-chlorohalopemide), which is a dual
inhibitor of PLD-1 and PLD-2 generation of PA [43],
counteracted the anti-proliferative/viability effect of D-
propranolol (Fig. 3D), thus indicating dependence of
D-propranolol effects on PLD-generated PA. These
results by themselves suggest that perturbing EGFR

FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS 2175

Targeting endocytosis blocks oncogenic EGFR R. Shaughnessy et al.

A B

C D

Fig. 2. Propranolol reduces EGFR degradation and prolongs its signaling activity. HeLa cells were incubated with EGF (50 ng
mL 1) in the
absence or presence of propranolol (100 lM) for the indicated times. (A) The mass of EGFR, assessed by immunoblot using the anti-EGFR-
968, decreases gradually to about 30% and propranolol prevents this decrease indicating a lower rate of degradation. (B) EGFR tyrosine
phosphorylation was analyzed by immunoblot with the monoclonal 4G10 anti-phosphotyrosine IgG after immunoprecipitation of the receptor
with the HB8506 antibody. Propranolol prolongs the lifetime of EGFR tyrosine phosphorylation that reflects an active receptor. (C)
Representative images of EGFR in early and recycling endosomes. After 1 h treatment with 50 ng
mL 1 EGF in the absence or presence of
75 lM propranolol, HeLa cells were fixed and stained with anti-EGFR and anti-EEA1 or anti-TfR along with Hoescht. In control conditions the
receptor localizes mainly at the cell surface, whereas it is internalized upon EGF or EGF and propranolol incubation. (D) Quantified
colocalization with EEA1 and TfR, (*P < 0.05, **P < 0.01 Student’s t test).

2176 FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS

R. Shaughnessy et al. Targeting endocytosis blocks oncogenic EGFR

A B C

D E F

Fig. 3. Acute/intermittent PAP inhibition selectively reduces EGFR-driven proliferation/viability by endocytic- and PA-dependent mechanisms.
(A) Permanently transfected HeLa cells overexpressing EGFR (HeLa-EGFR) show higher EGFR mass, assessed by immunoblot with the
EGFR-968 antibody (the slower migrating band corresponds to the EGFR-GFP) and higher proliferation rates compared with their parental
HeLa cells (ATCC). Mean  SEM of three experiments, *P < 0.001, Student’s t-test. (B) 125EGF binding shows similar EGFR surface levels
in HeLa (in-house) and permanently transfected HeLa-EGFR cells. (C) D-propranolol (75 lM) treatment for 20 min increased PA levels by
~ 60% in HeLa-EGFR cells. Mean  SEM of three experiments. *P < 0.05, Student’s t-test. (D) Effect of PAP inhibition with D-propranolol
on proliferation/viability of the indicated HeLa cells. Cells were incubated for 1 h, twice a day, each 6 h, during three consecutive days with
75 lM propranolol. After each treatment, medium was replaced with fresh complete medium. HeLa-EGFR cells were also treated in the
absence or presence of the PLD inhibitor FIPI (750 nM). The number of live cells was assessed counting cells that exclude trypan blue.
Mean  SEM of three experiments. n.s., not significant; *P < 0.05, **P < 0.0005, one-way ANOVA for indicated comparisons. (E) EGFR
internalization induced by propranolol (75 lM) treatment for 1 h is inhibited by co-incubation with 30 lM rolipram. Indirect
immunofluorescence was performed on fixed and permeabilized cells with anti-GFP (bar 10 lm). (F) The anti-proliferative/viability effect of
D-propranolol is counteracted by 30 lM rolipram in the acute/intermittent treatment described in (D). Mean  SEM of two experiments.
*P < 0.05, **P < 0.005, one-way ANOVA for indicated comparisons.

FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS 2177

Targeting endocytosis blocks oncogenic EGFR
endocytic trafficking by PAP inhibition can selectively
exert deleterious actions upon EGFR-dependent tu-
moral cells.
In our acute/intermittent treatment approach during
three consecutive days, conditions currently used to
inhibit endocytosis such as clathrin silencing with
small interfering RNA, cytosol acidification, sucrose
hyperosmotic treatment and dynosore all reduced pro-
liferation by 50% or more (not shown), precluding
their use to assess the role of endocytosis in the anti-
proliferation/viability effects of D-propranolol. How-
ever, we could directly test the role of endocytosis
using the specific PDE4 inhibitor rolipram, which we
have previously shown to inhibit EGFR internaliza-
tion induced by propranolol [23]. Rolipram treatment
for 1 h effectively abrogated endocytosis in HeLa-
EGFR cells (Fig. 3E), and used in the acute/intermit-
tent setting effectively decreased the anti-proliferative/
viability effect of D-propranolol (Fig. 3F). Therefore,
endocytosis seems to be required for the effect of
PAP inhibition on EGFR-dependent proliferation/via-
bility.
Taken together with the previous results we can con-
clude that an exaggerated PA signaling, such as that
triggered by D-propranolol, has the potential to coun-
teract the oncogenic influence of EGFR in cancerous
cells, involving endocytic traffic perturbations.
D-propranolol increases the pool of tyrosine-
phosphorylated EGFR in recycling endosomes
In the previous experiments we showed that PAP inhi-
bition with propranolol decreased the rate of degrada-
tion of total EGFR and increased the lifetime of
tyrosine-phosphorylated EGFR (Fig. 2A). The high
sensitivity of HeLa-EGFR cell proliferation/viability
to PAP inhibition prompted us to perform a more
detailed characterization of the endosomal distribution
of total and tyrosine-phosphorylated EGFR. After
60 min of EGF (50 ng
mL 1) treatment in the pres-
ence of 75 lM propranolol, we detected an increment
in the percentage of EGFR colocalizing with both
EEA1 and TfR positive endosomes, mainly distributed
in a more perinuclear region than in control cells with
EGF alone (Fig. 4A). The intracellular levels of tyro-
sine-phosphorylated EGFR (pEGFR), which reflect
ligand-activated EGFR, also increased and distributed
mainly to perinuclear endosomes in propranolol trea-
ted cells (Fig. 4B). Activated pEGFR accumulated in
perinuclear recycling endosomes presumably due to
recycling inhibition, as previously shown for empty/
inactive EGFR and TfR [23]. In addition, the distribu-
tion of pEGFR with respect to early (EEA1), recycling
2178
R. Shaughnessy et al.
(TfR) and late (Lamp1) endosomes showed a higher
increase in recycling endosomes (Fig. 4C). This obser-
vation agrees with the previous results showing
decreased EGFR degradation (Figs 2A and 4B), indi-
cating that propranolol diverts activated EGFR from
the lysosomal degradation route and lowers its recy-
cling. As suggested by the rolipram counteracting
effects, such endocytic perturbation seems to be
related to the anti-proliferative/viability effects of pro-
pranolol.
Prolonged treatment with lower concentrations
of D-propranolol and desipramine selectively
counteracts EGFR-dependent tumoral cell growth
We also studied the effects of continuous and pro-
longed treatments (72 h) with 75 lM D-propranolol,
which was indiscriminately toxic to our model cells,
including MDCK cells and our ‘in-house’ HeLa (not
shown). Therefore, we searched for milder treatment
conditions by testing combinations of concentrations
more closely related to those reported in the blood of
patients treated with D-propranolol and desipramine,
which would be safe for future trials in animal models.
D-propranolol has been used experimentally in man to
treat arrhythmia independently of b-adrenergic block-
age and has been found at concentrations up to 10 lM
measured in the blood several hours (twice a half-life)
after the last dose [44]. Desipramine reaches around
1 lM in antidepressive treatments [45–47]. Therefore,
we tested the effects of 10–30 lM D-propranolol alone
and in combination with 1 lM desipramine. At these
concentrations, D-propranolol still induced 10–20%
EGFR internalization in 30 min experiments, whereas
1 lM desipramine produced negligible effects in this
assay (Fig. 1A). However, these conditions again were
found to be highly selective against cells whose prolif-
eration strongly depends on the EGFR. For instance,
the proliferation/viability of HeLa-EGFR significantly
decreased with a combination of 30 lM D-propranolol
and 1 lM desipramine, contrasting with almost no
effect on their parental untransfected HeLa (ATCC)
cells (Fig. 5A). Furthermore, both drugs alone or in
combination were rather ineffective upon our ‘in-house’
HeLa cells (not shown), which as mentioned express
similar levels of EGFR to the transfected HeLa-EGFR.
MDCK cells were also not affected (Fig. 5A). The
recently acquired high levels of EGFR expression of
the transfected cells seemed to provide a cellular con-
text that promotes sensitivity to PAP inhibition.
Prolonged treatments of 24 or 48 h did not pro-
duce detectable changes in EGFR distribution
towards endosomal compartments (Fig. 5C). The
FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS
R. Shaughnessy et al. Targeting endocytosis blocks oncogenic EGFR

Fig. 4. PAP inhibition by propranolol increases the pool of internalized ligand-activated EGFR. HeLa-EGFR cells seeded on glass coverslips
were incubated with 50 ng
mL 1 EGF in the absence or presence of 75 lM propranolol for 30 and 60 min at 37 °C. The cells were then
fixed and co-stained with anti-GFP for total EGFR, anti-phospho-EGFR, to detect activated EGFR (pEGFR) and anti-EEA1, anti-TfR or anti-
Lamp1 for early, recycling and late endosomal localization, respectively. (A) Propranolol increases the percentage of total EGFR colocalizing
with the endosomal markers EEA1 and TfR, as shown by the quantitative analysis depicted in the graph. (B) The amount of intracellular
total EGFR and pEGFR increases in the presence of propranolol. (C) Mander’s coefficient indicates that propranolol (75 lM) increases the
amount of ligand-activated EGFR (pEGFR) in recycling endosomes and lowers it in late endosomes after 1 h treatment.

FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS 2179

Targeting endocytosis blocks oncogenic EGFR R. Shaughnessy et al.

A B

C D

Fig. 5. Continuous treatments with D-propranolol and desipramine in conditions that selectively reduce cell proliferation/viability of HeLa-
EGFR cells. (A) Proliferation of HeLa-wt (ATCC), HeLa-EGFR and (B) non-tumorigenic MDCK cells was assessed after 3 days under the
indicated treatments, which include the EGFR tyrosine kinase inhibitor AG1478 (0.1 lM), D-propranolol (D-prop) (10 and 30 lM) or
desipramine (DMI) (1 lM). While HeLa cells and MDCK cells are relatively insensitive to these drugs, HeLa-EGFR cell proliferation was
reduced especially by combined D-propranolol and desipramine treatment. Mean  SEM of three experiments. *P < 0.05; Student’s t-test.
(C) Immunofluorescence does not detect EGFR redistribution after treatments with the drugs for the indicated times. (D) EGFR mass is
maintained after prolonged/continuous treatments. Immunoblot was performed using anti-EGFR-968 after the indicated times of treatment
with 30 lM D-propranolol (P) and 1 lM desipramine (D).

EGFR mass showed no significant differences
between untreated or treated cells, indicating that the
deleterious effects of prolonged treatment with 30 lM
D-propranolol and 1 lM desipramine cannot be
explained by EGFR degradation (Fig. 5D). Subtle
alterations might still occur affecting EGFR endocytic
trafficking, yet undetectable to our analysis. Indeed, it
cannot be discarded that other deleterious mechanism
(s) might become preponderant under continuous
treatment with these lower concentrations of PAP
inhibitors. However, the similarity of the selectiveness
related to EGFR-driven growth seen with the acute/
intermittent treatment is suggestive of at least par-
tially shared mechanisms. Actually, as mentioned,
30 lM D-propranolol produced measurable endocyto-
sis of EGFR within 30 min (Fig. 1A) and this in
combination with desipramine might have effects sim-
ilar to those seen in acute/intermittent treatments.
Desipramine probably contributes to the anti-prolifer-
ation/viability effects by other mechanisms, as sug-
gested by the high sensitivity of other model cells (see
below).

2180 FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS

R. Shaughnessy et al. Targeting endocytosis blocks oncogenic EGFR

PAP inhibitors induce apoptosis and inhibit and desipramine compared with wild-type MKN45
anchorage-independent growth cells (Fig. 7A,C, respectively). These cells also internal-
ize EGFR under D-propranolol treatment (Fig. 7B).
We further characterized the effects of PAP inhibition
These results are consistent with the possibility that
on HeLa-EGFR cells, as these cells displayed selective
overexpression of EGFR as an acquired tumorigenic
sensitivity to PAP inhibition compared with their
trait increases the sensitivity to PAP inhibitors. Inter-
parental HeLa (ATCC) cells. Immunoblots against
estingly, MKN-EGFR cells were particular sensitive to
PARP, a substrate that is cleaved by activated cas-
1 lM desipramine alone, contrasting with the HeLa-
pase-3 [48], showed evidence of apoptosis induced by
EGFR cells. We do not yet know the source of such a
the combination of 30 lM D-propranolol and 1 lM
difference but very probably it involves another mech-
desipramine (Fig. 6A). Furthermore, D-propranolol
anism distinct from endocytic perturbation. The com-
also reduced the capability of HeLa-EGFR cells to
bined effects of D-propranolol and desipramine are
form colonies in soft agar (Fig. 6B), which reflects
higher in these cells.
anchorage-independent growth considered to be a hall-
We then tested these same treatment conditions on
mark of malignancy [49].
cells that have naturally acquired oncogenic EGFR
function. To this end we used widely characterized
D-propranol and desipramine are also effective H1975 cells from non-small-cell lung cancer cells,
against other EGFR-dependent cancerous cells which express a double mutant EGFRL858R/T790M and
are highly oncogenic and resistant to current receptor
To analyze how extensive the effects of continuous
tyrosine kinase inhibitors [51,52]. Binding assays and
treatment with D-propranolol and desipramine are to
immunofluorescence showed that these cells internalize
other model cells, we produced permanent transfec-
EGFR in response to propranolol (Fig. 7D,E). Strik-
tants of EGFR in the gastric cancer cell line MKN45,
ingly, their proliferation/viability decreased under
which do not express high levels of EGFR at the cell
prolonged treatment with a combination of 30 lM
surface [50]. Similar to HeLa-EGFR cells, MKN45-
D-propranolol and 1 lM desipramine (Fig. 7F).
EGFR cells showed increased expression of EGFR
Altogether, these results indicate that treatment with
(Fig. 7A, inset) that translated into a higher rate of
PAP inhibitors can be adjusted to selectively damage
proliferation and higher sensitivity to D-propranolol

A B

Fig. 6. PAP inhibitors induce apoptosis and reduce anchorage-independent growth in HeLa-EGFR cells. (A) Immunoblot analysis of apoptosis
in HeLa-EGFR cells treated with staurosporine (positive control), or a combination of D-propranolol (D-prop) (30 lM) and desipramine (DMI)
(1 lM) for the indicated time points, in the presence of 10% fetal bovine serum. Medium was replaced every 24 h. A polyclonal antibody
that detects the full-length PARP protein and its 89 kDa cleaved fragment produced by caspase-3 activity indicates increased apoptosis after
48 h of treatment. Numbers are the percentage of cleaved versus full-length PARP. b-actin was used as a loading control. (B) Anchorage-
independent growth of HeLa-EGFR cells was reduced by D-propranolol. Cells were seeded in 12-well plates containing soft agar and after
24 h were either treated with D-propranolol (25 lM) or left untreated for 14 days. Formed colonies counted manually show a decreased
number under D-propranolol treatment.

FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS 2181

Targeting endocytosis blocks oncogenic EGFR R. Shaughnessy et al.

A B C

D E F

Fig. 7. EGFR-dependent proliferation/viability of MKN45-EGFR and H1975 cells is sensitive to prolonged/continuous treatment with
D-propranolol and desipramine. (A) Gastric cancer MKN45 cells permanently transfected to overexpress the EGFR show higher levels of

EGFR in immunoblot with EGFR-968 antibody (inset) and higher proliferation rates than their parental MKN45 cells. (B) D-propranolol (D-prop)
(75 lM) induces internalization of MKN45-EGFR cells, as shown by GFP fluorescence. (C) D-propranolol in combination with desipramine
selectively decreases proliferation/viability of MKN-EGFR cells, contrasting with parental MKN45 cells. MKN45 cells also display higher
sensitivity to the EGFR tyrosine kinase inhibitor AG1478 (0.1 lM). Mean  SEM of three experiments. *P < 0.001, one-way ANOVA for
indicated comparisons with control treatment. (D) Lung cancer cells H1975 expressing a double mutant EGFR, which is both oncogenic and
resistant to current tyrosine kinase activity, internalize the receptor in response to either propranolol (75 lM) or desipramine (20 lM) or both.
(E) H1975 immunofluorescence shows EGFR endocytosis under D-propranolol (D-prop) treatment. (F) Proliferation/viability of H1975 cells is
sensitive to inhibition by D-propranolol (D-prop; 30 lM) alone or in combination with desipramine (DMI; 1 lM). *P < 0.05, **P < 0.005 one-
way ANOVA for indicated comparisons with control treatment.

2182 FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS

R. Shaughnessy et al.
cancerous cells whose proliferation depends on EGFR
overexpression or activating mutations as acquired
oncogenic traits.
Discussion
This work provides novel evidence to consider PAP
activity as a suitable target against cancer. In contrast
to current drugs that directly target the EGFR mole-
cule, PAP inhibitors have the potential to counteract
EGFR-dependent cell proliferation/viability of cancer-
ous cells through a mechanism involving PA-mediated
endocytic regulation. Our results indicate that PAP
inhibition can selectively reduce cell proliferation and
induce apoptosis in tumoral cells that have acquired
oncogenic EGFR dysfunction by overexpression or
oncogenic mutation. PAP inhibition makes unoccu-
pied EGFR inaccessible to stimulation due to inter-
nalization and prolongs the lifetime of ligand-
activated EGFRs at recycling endosomes, diverging
them from the degradation pathway. Cell prolifera-
tion/viability becomes impaired by PAP inhibition,
unless EGFR endocytosis or PLD generation of PA is
concomitantly blocked. Therefore, endocytic traffick-
ing perturbations derived from exaggerated PA signal-
ing can have deleterious effects upon EGFR-
dependent tumoral cells. Besides their relevance in
cancer, these results also highlight the role of endocy-
tic trafficking as a critical modulator of EGFR-medi-
ated cell responses.
PLD-generated PA has previously been proposed as
a target to counteract cancerous behavior but focusing
on the inhibition of PLD activity [33,53–56]. PLD
activity is increased in response to mitogenic signals
and activated oncogenes, and its expression is fre-
quently found to be elevated in cancerous cells, seem-
ingly contributing to cell transformation, tumoral
growth and malignancy [54,55]. PLD overproduction
of PA might contribute to cell malignancy by increas-
ing the activity of the MAPK pathway through PA-
mediated recruitment of SOS and Raf [56–60] and also
by PAP-mediated rapid generation of DAG [54], all
important elements in cell proliferation and cancer
[61,62]. Accordingly, inhibitors of PLD have been aris-
ing and have proved to be effective against tumoral
cell proliferation [33,53], migration and metastasis [55].
Thus, the alternative of inhibiting PAP activity to
exaggerate PA signaling indeed appears at odds with
the proposed role of PLD-generated PA in cancer.
However, our present evidence strongly suggests that
acute increments in PA signaling might be beneficial at
least in the context of oncogenic EGFR dysfunction,
acting through endocytic traffic perturbation.
FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS
Targeting endocytosis blocks oncogenic EGFR
Previous work [23] showed that propranolol used as
a PAP inhibitor induces endocytosis of empty/inactive
EGFR through both clathrin-dependent and clathrin-
independent pathways, and also decreases recycling of
the receptor. As a result of these two perturbations the
EGFR accumulates mainly at recycling endosomes, in
a reversible manner, returning to the cell surface after
discontinuing the treatment. The mechanism involves a
signaling pathway initiated by transient PA elevations
that activates PDE4 and decreases cAMP and PKA
activity. Thus, EGFR internalization induced by pro-
pranolol can be effectively reduced by inhibiting PLD
with FIPI or PDE4 with rolipram [23]. Here we add
that propranolol reduces the degradation of ligand-
activated EGFRs and increases their pool at recycling
endosomes, mimicking previously reported effects of
PKA inhibitors [24].
Endosomal accumulation of activated EGFRs
leading to apoptosis has recently been described in
MDA-MB-468 cells expressing extremely high levels of
EGFRs (around 106 EGFRs
cell 1), which saturate
the endocytic machinery, as well as in HeLa cells
expressing physiological levels of EGFR and treated
with monensin to accumulate activated EGFR in early
endosomes [20]. Because PAP inhibition delays degra-
dation and increases the endosomal pool of ligand-
activated EGFR, while empty/inactive EGFRs can
undergo internalization and thus become inaccessible
to further stimulation, it seems likely that these two
endocytic perturbations contribute to lower EGFR-
dependent cell proliferation. Such a possibility is sup-
ported by the effect of rolipram, which inhibits EGFR
endocytosis and the decrease in proliferation/viability
induced by propranolol.
We show that L- and D-propranolol, which are equiv-
alent in PDE4 activation derived from PAP inhibition
[27], are also equivalent in inducing EGFR internaliza-
tion. Desipramine is a more potent PAP inhibitor than
propranolol [37] and also triggers the PA/PDE4/PKA
pathway [27]. Correlating with these properties, we
show that desipramine induces EGFR internalization
displaying lower EC50 (20 lM) than propranolol
(75 lM). Propranolol and desipramine used at their cor-
responding EC50, far below the concentrations currently
used to completely inhibit PAP activity [40,41], have
additive effects on EGFR internalization. Therefore,
propranolol or its isomers can be used interchangeably
as PAP inhibitors, either alone or in combination with
desipramine, to perturb EGFR endocytosis.
Oncogenic signaling due to overexpression or muta-
tion of EGFR is frequently found in epithelial
cell derived cancers [1,3] and is currently targeted in
anti-tumoral therapies, expecting to produce less delete-
2183
Targeting endocytosis blocks oncogenic EGFR
rious effects to normal cells than conventional chemo-
therapy [2,3]. PAP inhibition might meet the criterion
of selectivity against EGFR-dependent cancer traits.
Our permanently transfected model cells resemble an
acquired trait of EGFR overexpression leading to
increased proliferation. Strikingly, intermittent treat-
ments with D-propranolol at the EC50 for EGFR inter-
nalization, applied twice a day for 1 h and leaving
intervals of 6 h without the drug, is enough to decrease
proliferation/viability in HeLa-EGFR, but not in
HeLa-wt cells or ‘in-house’ HeLa cells expressing
similar EGFR levels. MDCK cells exemplifying normal
epithelial cells [42] are also resistant to such treatment
(not shown). It seems that recently acquired overexpres-
sion of EGFR provides sensitivity to PAP inhibitors.
To test the role of endocytosis we used rolipram,
which blocks both clathrin-dependent and clathrin-
independent endocytosis of EGFR internalization
induced by propranolol [23]. We show that rolipram
alone does not affect cell proliferation in the acute/
intermittent treatment protocol, but inhibits EGFR
internalization and the anti-proliferative/viability
effects of D-propranolol. Furthermore, inhibition of
PLD activity with FIPI [43] also reduces such effects,
demonstrating their dependence on PLD-generated
PA. All these results indicate that pharmacological
perturbations or EGFR endocytic trafficking derived
from exaggerated PA signaling have the potential to
selectively counteract the oncogenic influence of
EGFR in cancerous cells.
PAP inhibition reduces PA-generated DAG [40,41],
which depending on its magnitude might decrease
DAG-mediated signaling and lipid homeostasis [63].
Propranolol at twice the concentration used here has
been shown to induce apoptosis in numerous cell lines
associated with impairments in lipid homeostasis [63].
Such general deleterious alteration should indiscrimi-
nately affect all kinds of cells, contrasting with the
selective cellular sensitivity shown here, which relates
to EGFR oncogenic function. Alterations in lipid
homeostasis might contribute together with EGFR
endocytic trafficking perturbations to determine cell-
context-dependent deleterious effects.
Our results also show interesting results using com-
binations of D-propranolol and desipramine at concen-
trations closely related to those reported in the blood
of patients treated with these drugs [44,46]. HeLa-
EGFR and MKN-EGFR cells, which have increased
proliferation compared with their parental counter-
parts, display higher sensitivity to continuous treat-
ment (72 h) with 10–30 lM D-propranolol and 1 lM
desipramine, while epithelial MDCK cells remain unaf-
fected with this treatment. H1975 lung cancer cells that
2184
R. Shaughnessy et al.
express a double mutation of EGFR, which is not only
highly oncogenic but also resistant to current receptor
tyrosine kinase inhibitors [51,52], are also highly sensi-
tive. As shown in HeLa-EGFR cells, this prolonged
low dose treatment can also induce apoptosis and
reduce anchorage-independent growth, a hallmark of
malignancy [49]. However, after 24 or 48 h under this
treatment, we could not detect changes in EGFR cellu-
lar distribution and the mass of EGFRs is maintained,
discarding downregulation as a deleterious mechanism.
The concentration-dependent analysis shows that D-
propranolol at 10–30 lM induces 15–20% EGFR inter-
nalization within 30 min (see Fig. 1A), while at 1 lM
desipramine endocytosis is almost negligible. Because
we noticed that daily changes in medium with fresh
drug are necessary to see the anti-proliferation/viability
effects, it is possible that D-propranolol contributes to
reduce proliferation/viability through acute endocytic
perturbations similar to those seen under acute/inter-
mittent treatments. The high sensitivity of MKN45
cells to 1 lM desipramine alone suggests a different yet
unknown mechanism, which might operate in combi-
nation with D-propranolol. Subtle changes in the endo-
cytic trafficking of the EGFR might still occur,
undetectable by our analysis at steady state conditions
but enough to cause an imbalance in EGFR signaling
towards deleterious effects.
PAP activity involves two different enzyme families,
PAP1 and PAP2, each comprising three isoforms [34].
These include soluble lipins-1–3 (PAP1), which associ-
ate with membranes of the endoplasmic reticulum, and
transmembrane lipid phosphate phosphatases LPP-1–3
(PAP2), which are distributed at the plasma membrane
and endosomal membranes and have a broader range
of substrates beyond PA. Lipin-1 knockdown cells
contain increased levels of PA and decreased levels of
cAMP and PKA activity, suggesting that soluble PAP
activity regulates the levels of PA that modulate the
PDE4/PKA pathway [64]. There are also examples of
cellular effects of propranolol acting mainly upon solu-
ble PAP activity [37,40]. Lipins are thus likely candi-
dates to mediate the effects of PAP inhibitors on
EGFR endocytic traffic perturbation, remaining to be
experimentally proved.
Because PA is involved in a variety of cellular pro-
cesses, acting through a large number of PA binding
effectors [29], more than a single mechanism might be
involved in the anti-proliferation/viability effects of
PAP inhibition. Spatiotemporal alterations in EGFR
signaling might combine with other PA actions to be
disclosed. These might include defects in DAG produc-
tion leading to unbalanced lipid homeostasis [63] or
alterations in macroautophagy recently involved in
FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS
R. Shaughnessy et al.
EGFR oncogenic function [65]. However, of all these
possibilities, the inhibitory effect of rolipram down-
stream of PLD-generated PA puts forward a higher
role of lowering cAMP levels and PKA activity to
induce EGFR endocytic perturbations in mediating
anti-tumoral effects of PAP inhibition. Desipramine
used at high concentrations, enough to trigger the PA/
PDE4/PKA pathway, has been shown to induce apop-
tosis in several carcinoma cells, so far attributed to the
activation of JNK [66], endoplasmic reticulum stress
[67] and autophagy [68], although without analyzing
EGFR internalization. Whatever the complexity of the
mechanisms involved, PAP inhibition arises as an
alternative to selectively induce deleterious effects
upon cancerous cells whose malignancy depends on
oncogenic EGFR dysfunctions.
Materials and methods
Antibodies and reagents
Anti-phosphotyrosine monoclonal IgG and anti-pEG-
FRY1068 (Millipore, Billerica, MA, USA), anti-EGFR-968
[42], monoclonal anti-EGFR HB8506 (Hybridoma) (ATCC,
Manassas, VA, USA), polyclonal anti-poly ADP ribose
polymerase (PARP) (Cell Signaling Technology, Danvers,
MA, USA), rabbit polyclonal anti-GFP (produced in our
laboratory), anti-EEA1 and anti-TfR (BD Transduction
Laboratory, Franklin Lakes, NJ, USA), anti-Lamp1 (Santa
Cruz, Palo Alto, CA, USA) and anti-b-actin (Abcam,
Cambridge, UK) were used. The ECL system (Thermo
Fisher Scientific, Waltham, MA, USA) was used for
immunoblot detection. Secondary antibodies were Alexa 488
and Alexa 555 (Molecular Probes, Carlsbad, CA, USA) and
HRP-conjugated goat anti-rabbit and anti-mouse IgG
(Rockland Immunochemicals, Boyertown, PA, USA). D-pro-
pranolol, L-propranolol, desipramine, agarose type II and
bovine serum albumin (BSA) from Sigma-Aldrich, St Louis,
MO, USA. Plasmid pEGFP-N1-EGFR for EGFR-GFP
expression was provided by Alexander Sorkin (University of
Pittsburgh, PA, USA) [69]. Cell culture reagents were pur-
chased from Invitrogen (Carlsbad, CA, USA) and Sigma
Aldrich. EGFR and PLD inhibitors, AG1478 and FIPI, were
used (Calbiochem, Darmstadt, Germany). Tissue culture
plastics were purchased from Nalgene Nunc (Thermo Fisher
Scientific).
Cell culture, transfections and proliferation
assays
HeLa cells (ATCC) and MKN45 cells (JCRB0254, from
Andrew Quest, Universidad de Chile, Santiago, Chile) were
permanently transfected with pEGFP-N1-EGFR (HeLa-EGFR
or MKN45-EGFR) using the Lipofectamine 2000 method,
FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS
Targeting endocytosis blocks oncogenic EGFR
selected with 1 mg
mL 1 geneticin sulfate (G418) and main-
tained in 0.8 mg
mL 1 G418. Cultures of HeLa cells and
H1975 (ATCC, from K. Polity, Yale University, USA) were
in DMEM and MKN45 cells in RPMI, supplemented with
10% fetal bovine serum and antibiotics (200 U
mL 1 penicil-
lin and 0.1 mg
mL 1 streptomycin; Sigma-Aldrich), kept at
37 °C in a humidified atmosphere (95% air, 5% CO2). Cell
growth was analyzed by counting trypan blue stained cells
(Invitrogen) at a ratio of 1 : 1 using the Countess Auto-
matic Cell Counter (Invitrogen). Cells were seeded at 10 000
cells
well 1 and drug-containing medium was replaced every
24 h. Data were graphed and analyzed using the Student’s
t test using GRAPHPAD PRISM software (San Diego, CA, USA).
PA determination
Cellular PA levels were measured using an enzyme-coupled
fluorometric assay, as previously described [70]. Fluores-
cence emission of resorufin reporter was measured at
580 nm after excitation at 530 nm in a Synergy fluorometer
(Biotek, Winooski, VT, USA) using GEN5 2.0 ALL-IN-ONE MI-
CROPLATE READER software (Biotek). The PA concentration
was standardized to total protein concentration.
Ligand-binding assays
Human 125I-EGF was prepared by the chloramine T
method obtaining specific activities of 50 000–70 000
cpm
ng 1 and binding assays were done as described previ-
ously [23,24] in Hanks’ solution with 20 mM HEPES and
0.2% BSA during 1 h at 4 °C. After washing the unbound
ligand, the cells were lysed with 1 M NaOH. Total counts
per minute was measured in a Wizard2 Automatic Gamma
Counter (PerkinElmer, Waltham, MA, USA). The samples
were normalized by total amount of proteins measured
spectrophotometrically at 280 nm.
Immunoprecipitation and immunoblot assays
Extracts of 250 lg of total protein was immunoprecipitated
with the monoclonal EGFR-HB8506 antibody, resolved by
SDS/PAGE and immunoblotted with anti-phosphotyrosine
monoclonal IgG (Millipore) as previously described [24].
To detect EGFR, PARP (Cell Signaling) and b-actin (Ab-
cam) cell extracts were subjected to 7.5% SDS/PAGE and
the proteins were electrotransferred onto nitrocellulose
membranes and immunoblotted with the corresponding
antibodies. Anti-EGFR-968 was used for EGFR [42].
Bands were revealed by ECL (Thermo Fisher Scientific)
and visualized with the Chemidoc IT system using the VI-
SIONWORKS LS software. Digital pictures were analyzed with
IMAGEJ (NIH, Bethesda, MD, USA) and Adobe PHOTOSHOP
CS (Adobe Systems, Mountain View, CA, USA). For
assessing apoptosis we used anti-PARP (Cell Signaling),
2185
Targeting endocytosis blocks oncogenic EGFR
which detects both full-length PARP and its 89 kDa
cleaved fragment produced by activated caspase 3 [48].
Western blot quantification was performed using IMAGEJ
software and graphed using GRAPHPAD PRISM software as a
percentage of cleaved PARP/total PARP.
Immunofluorescence analysis
Cells grown on glass coverslips were prepared for immuno-
fluorescence as described previously [71]. The secondary
antibodies were anti-mouse AF555 and anti-rabbit AF488
[71]. Cells were then mounted on microscope slides with
Fluoromount G and digital fluorescence images were
acquired on a Leica DMI600b inverted microscope with a
639 glycerol immersion lens and an Andor iXon3 EMCCD
camera. Z stacks of 200 nm steps were acquired and then
3D deconvoluted using SVI HUYGENS ESSENTIAL SOFTWARE
(SVI, Hilversum, The Netherlands). Colocalization analysis
was performed using IMAGEJ and JACOP plugin. HeLa and
MKN45 cells stably expressing EGFR-GFP were fixed as
above and mounted on microscope slides with Fluoromount
G and digital direct fluorescence images were acquired on a
Zeiss Axiophot microscope with a Plan-APOCHROMAT
100/1.4 oil immersion objective and the 14-bit Zeiss Axio-
cam camera, transferred to a computer workstation running
AXIOVISION imaging software (Zeiss, Thornwood, NY,
USA), processed and analyzed with IMAGEJ software.
Soft agar colony formation assay
Assays were performed in 12-well plates with
300 cells
well 1, resuspended as a single cell suspension in
0.3% agar and layered on top of 0.5% agar, as described
previously [72]. The plates were incubated for 14 days and
formed colonies were counted manually under 109 objec-
tive lenses, using at least three wells per condition in each
experiment.
Acknowledgements
This work received financial support from grants
CONICYT # PFB12/2007, Fondecyt #1100747, FON-
DEF # D09I1104, Program MIFAB-FCHBC-MF105
and CONICYT fellowship to RS. We are grateful to
Alexander Sorkin for his generous gift of EGFR-GFP
expression plasmid and to Andrew Quest and Katerina
Polity for providing cell lines. The authors have no
conflicting financial interests.
References
1 Yarden Y & Pines G (2012) The ERBB network: at
last, cancer therapy meets systems biology. Nat Rev
Cancer 12, 553–563.
2186
R. Shaughnessy et al.
2 Ciardiello F & Tortora G (2008) EGFR antagonists in
cancer treatment. N Engl J Med 358, 1160–1174.
3 Mellinghoff I (2007) Why do cancer cells become
‘addicted’ to oncogenic epidermal growth factor
receptor? PLoS Med 4, 1620–1622.
4 Pao W & Chmielecki J (2010) Rational, biologically
based treatment of EGFR-mutant non-small-cell lung
cancer. Nat Rev Cancer 10, 760–774.
5 Goh LK & Sorkin A (2013) Endocytosis of receptor
tyrosine kinases. Cold Spring Harb Perspect Biol 5,
a017459.
6 Di Fiore PP & De Camilli P (2001) Endocytosis and
signaling: an inseparable partnership. Cell 106, 1–4.
7 Brankatschk B, Wichert SP, Johnson SD, Schaad O,
Rossner MJ & Gruenberg J (2012) Regulation of the
EGF transcriptional response by endocytic sorting.
Sci Signal 5, ra21.
8 Ceresa BP & Schmid SL (2000) Regulation of signal
transduction by endocytosis. Curr Opin Cell Biol 12,
204–210.
9 Miaczynska M, Pelkmans L & Zerial M (2004) Not just
a sink: endosomes in control of signal transduction.
Curr Opin Cell Biol 16, 400–406.
10 Goh LK, Huang F, Kim W, Gygi S & Sorkin A (2010)
Multiple mechanisms collectively regulate clathrin-
mediated endocytosis of the epidermal growth factor
receptor. J Cell Biol 189, 871–883.
11 Sigismund S, Algisi V, Nappo G, Conte A, Pascolutti
R, Cuomo A, Bonaldi T, Argenzio E, Verhoef LG,
Maspero E et al. (2013) Threshold-controlled
ubiquitination of the EGFR directs receptor fate.
EMBO J 32, 2087–2185.
12 Stang E, Blystad FD, Kazazic M, Bertelsen V, Brodahl
T, Raiborg C, Stenmark H & Madshus IH (2004) Cbl-
dependent ubiquitination is required for progression of
EGF receptors into clathrin-coated pits. Mol Biol Cell
15, 3591–3604.
13 Levkowitz G, Waterman H, Zamir E, Kam Z, Oved S,
Langdon WY, Beguinot L, Geiger B & Yarden Y
(1998) c-Cbl/Sli-1 regulates endocytic sorting and
ubiquitination of the epidermal growth factor receptor.
Genes Dev 12, 3663–3674.
14 Marmor MD & Yarden Y (2004) Role of protein
ubiquitylation in regulating endocytosis of receptor
tyrosine kinases. Oncogene 23, 2057–2070.
15 Sigismund S, Argenzio E, Tosoni D, Cavallaro E, Polo
S & Di Fiore PP (2008) Clathrin-mediated
internalization is essential for sustained EGFR
signaling but dispensable for degradation. Dev Cell 15,
209–219.
16 Wegner CS, Rodahl LM & Stenmark H (2011) ESCRT
proteins and cell signalling. Traffic 12, 1291–1297.
17 Di Fiore PP (2009) Endocytosis, signaling and cancer,
much more than meets the eye. Preface. Mol Oncol 3,
273–279.
FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS
R. Shaughnessy et al.
18 Mosesson Y, Mills GB & Yarden Y (2008) Derailed
endocytosis: an emerging feature of cancer. Nat Rev
Cancer 8, 835–850.
19 Bache KG, Slagsvold T & Stenmark H (2004) Defective
downregulation of receptor tyrosine kinases in cancer.
EMBO J 23, 2707–2712.
20 Rush JS, Quinalty LM, Engelman L, Sherry DM &
Ceresa BP (2012) Endosomal accumulation of the
activated epidermal growth factor receptor (EGFR)
induces apoptosis. J Biol Chem 287, 712–722.
21 Sorkin A & von Zastrow M (2009) Endocytosis and
signalling: intertwining molecular networks. Nat Rev
Mol Cell Biol 10, 609–622.
22 Pelkmans L, Fava E, Grabner H, Hannus M,
Habermann B, Krausz E & Zerial M (2005) Genome-
wide analysis of human kinases in clathrin- and
caveolae/raft-mediated endocytosis. Nature 436, 78–86.
23 Norambuena A, Metz C, Jung JE, Silva A, Otero C,
Cancino J, Retamal C, Valenzuela JC, Soza A &
Gonzalez A (2010) Phosphatidic acid induces ligand-
independent epidermal growth factor receptor endocytic
traffic through PDE4 activation. Mol Biol Cell 21,
2916–2929.
24 Salazar G & Gonzalez A (2002) Novel mechanism for
regulation of epidermal growth factor receptor
endocytosis revealed by protein kinase A inhibition.
Mol Biol Cell 13, 1677–1693.
25 Taylor SS, Kim C, Cheng CY, Brown SH, Wu J &
Kannan N (2008) Signaling through cAMP and cAMP-
dependent protein kinase: diverse strategies for drug
design. Biochim Biophys Acta 1784, 16–26.
26 Wettschureck N & Offermanns S (2005) Mammalian G
proteins and their cell type specific functions. Physiol
Rev 85, 1159–1204.
27 Grange M, Sette C, Cuomo M, Conti M, Lagarde M,
Prigent AF & Nemoz G (2000) The cAMP-specific
phosphodiesterase PDE4D3 is regulated by
phosphatidic acid binding. Consequences for cAMP
signaling pathway and characterization of a
phosphatidic acid binding site. J Biol Chem 275, 33379–
33387.
28 Grange M, Picq M, Prigent AF, Lagarde M & Nemoz
G (1998) Regulation of PDE-4 cAMP
phosphodiesterases by phosphatidic acid. Cell Biochem
Biophys 29, 1–17.
29 Liu Y, Su Y & Wang X (2013) Phosphatidic acid-
mediated signaling. Adv Exp Med Biol 991, 159–176.
30 Roth MG (2008) Molecular mechanisms of PLD
function in membrane traffic. Traffic 9, 1233–1239.
31 Antonescu CN, Danuser G & Schmid SL (2010)
Phosphatidic acid plays a regulatory role in clathrin-
mediated endocytosis. Mol Biol Cell 21, 2944–2952.
32 Shen Y, Xu L & Foster DA (2001) Role for
phospholipase D in receptor-mediated endocytosis. Mol
Cell Biol 21, 595–602.
FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS
Targeting endocytosis blocks oncogenic EGFR
33 Selvy PE, Lavieri RR, Lindsley CW & Brown HA
(2011) Phospholipase D: enzymology, functionality, and
chemical modulation. Chem Rev 111, 6064–6119.
34 Kok BP, Venkatraman G, Capatos D & Brindley DN
(2012) Unlike two peas in a pod: lipid phosphate
phosphatases and phosphatidate phosphatases. Chem
Rev 112, 5121–5146.
35 Houslay MD (2001) PDE4 cAMP-specific
phosphodiesterases. Prog Nucleic Acid Res Mol Biol 69,
249–315.
36 Blackman BE, Horner K, Heidmann J, Wang D,
Richter W, Rich TC & Conti M (2011) PDE4D and
PDE4B function in distinct subcellular compartments in
mouse embryonic fibroblasts. J Biol Chem 286, 12590–
12601.
37 Meier KE, Gause KC, Wisehart-Johnson AE, Gore
AC, Finley EL, Jones LG, Bradshaw CD, McNair AF
& Ella KM (1998) Effects of propranolol on
phosphatidate phosphohydrolase and mitogen-activated
protein kinase activities in A7r5 vascular smooth
muscle cells. Cell Signal 10, 415–426.
38 Alexander RW, Williams LT & Lefkowitz RJ (1975)
Identification of cardiac beta-adrenergic receptors by
(minus) [3H]alprenolol binding. Proc Natl Acad Sci
USA 72, 1564–1568.
39 Howe R & Shanks RG (1966) Optical isomers of
propranolol. Nature 210, 1336–1338.
40 Asp L, Kartberg F, Fernandez-Rodriguez J, Smedh M,
Elsner M, Laporte F, Barcena M, Jansen KA, Valentijn
JA, Koster AJ et al. (2009) Early stages of Golgi vesicle
and tubule formation require diacylglycerol. Mol Biol
Cell 20, 780–790.
41 Baron CL & Malhotra V (2002) Role of diacylglycerol
in PKD recruitment to the TGN and protein transport
to the plasma membrane. Science 295, 325–328.
42 Buvinic S, Bravo-Zehnder M, Boyer JL, Huidobro-
Toro JP & Gonzalez A (2007) Nucleotide P2Y1
receptor regulates EGF receptor mitogenic signaling
and expression in epithelial cells. J Cell Sci 120, 4289–
4301.
43 Su W, Yeku O, Olepu S, Genna A, Park JS, Ren H,
Du G, Gelb MH, Morris AJ & Frohman MA (2009) 5-
Fluoro-2-indolyl des-chlorohalopemide (FIPI), a
phospholipase D pharmacological inhibitor that alters
cell spreading and inhibits chemotaxis. Mol Pharmacol
75, 437–446.
44 Murray KT, Reilly C, Koshakji RP, Roden DM,
Lineberry MD, Wood AJ, Siddoway LA, Barbey JT &
Woosley RL (1990) Suppression of ventricular
arrhythmias in man by D-propranolol independent of
beta-adrenergic receptor blockade. J Clin Invest 85,
836–842.
45 Frazer A (2001) Serotonergic and noradrenergic
reuptake inhibitors: prediction of clinical effects from in
vitro potencies. J Clin Psychiatry 62 (Suppl 12), 16–23.
2187
Targeting endocytosis blocks oncogenic EGFR
46 Hursting MJ, Clark GD, Raisys VA, Miller SJ &
Opheim KE (1992) Measurement of free desipramine in
serum by ultrafiltration with immunoassay. Clin Chem
38, 2468–2471.
47 Bailey DN & Jatlow PI (1976) Gas-chromatographic
analysis for therapeutic concentration of imipramine
and disipramine in plasma, with use of a nitrogen
detector. Clin Chem 22, 1697–1701.
48 Nicholson DW, Ali A, Thornberry NA, Vaillancourt
JP, Ding CK, Gallant M, Gareau Y, Griffin PR,
Labelle M, Lazebnik YA et al. (1995) Identification
and inhibition of the ICE/CED-3 protease necessary for
mammalian apoptosis. Nature 376, 37–43.
49 Hanahan D & Weinberg RA (2011) Hallmarks of
cancer: the next generation. Cell 144, 646–674.
50 Hotz B, Keilholz U, Fusi A, Buhr HJ & Hotz HG
(2012) In vitro and in vivo antitumor activity of
cetuximab in human gastric cancer cell lines in relation
to epidermal growth factor receptor (EGFR) expression
and mutational phenotype. Gastric Cancer 15, 252–264.
51 Pao W, Miller VA, Politi KA, Riely GJ, Somwar R,
Zakowski MF, Kris MG & Varmus H (2005) Acquired
resistance of lung adenocarcinomas to gefitinib or
erlotinib is associated with a second mutation in the
EGFR kinase domain. PLoS Med 2, e73.
52 Ohashi K, Maruvka YE, Michor F & Pao W (2013)
Epidermal growth factor receptor tyrosine kinase
inhibitor-resistant disease. J Clin Oncol 31, 1070–1080.
53 Scott SA, Selvy PE, Buck JR, Cho HP, Criswell TL,
Thomas AL, Armstrong MD, Arteaga CL, Lindsley
CW & Brown HA (2009) Design of isoform-selective
phospholipase D inhibitors that modulate cancer cell
invasiveness. Nat Chem Biol 5, 108–117.
54 Foster DA (2009) Phosphatidic acid signaling to
mTOR: signals for the survival of human cancer cells.
Biochim Biophys Acta 1791, 949–955.
55 Su W, Chen Q & Frohman MA (2009) Targeting
phospholipase D with small-molecule inhibitors as a
potential therapeutic approach for cancer metastasis.
Future Oncol 5, 1477–1486.
56 Zhang F, Wang Z, Lu M, Yonekubo Y, Liang X,
Zhang Y, Wu P, Zhou Y, Grinstein S, Hancock JF
et al. (2014) Temporal production of the signaling lipid
phosphatidic acid by phospholipase d2 determines the
output of extracellular signal-regulated kinase signaling
in cancer cells. Mol Cell Biol 34, 84–95.
57 Ghosh S, Strum JC, Sciorra VA, Daniel L & Bell RM
(1996) Raf-1 kinase possesses distinct binding domains
for phosphatidylserine and phosphatidic acid.
Phosphatidic acid regulates the translocation of Raf-1
in 12-O-tetradecanoylphorbol-13-acetate-stimulated
Madin Darby canine kidney cells. J Biol Chem 271,
8472–8480.
58 Andresen BT, Rizzo MA, Shome K & Romero G
(2002) The role of phosphatidic acid in the regulation
2188
R. Shaughnessy et al.
of the Ras/MEK/Erk signaling cascade. FEBS Lett 531,
65–68.
59 Lee CS, Kim KL, Jang JH, Choi YS, Suh PG &
Ryu SH (2009) The roles of phospholipase D in
EGFR signaling. Biochim Biophys Acta 1791, 862–
868.
60 Zhang Y & Du G (2009) Phosphatidic acid signaling
regulation of Ras superfamily of small guanosine
triphosphatases. Biochim Biophys Acta 1791, 850–855.
61 Santarpia L, Lippman SM & El-Naggar AK (2012)
Targeting the MAPK-RAS-RAF signaling pathway in
cancer therapy. Expert Opin Ther Targets 16, 103–
119.
62 Griner EM & Kazanietz MG (2007) Protein kinase C
and other diacylglycerol effectors in cancer. Nat Rev
Cancer 7, 281–294.
63 Fuentes L, Perez R, Nieto ML, Balsinde J & Balboa
MA (2003) Bromoenol lactone promotes cell death by a
mechanism involving phosphatidate phosphohydrolase-
1 rather than calcium-independent phospholipase A2. J
Biol Chem 278, 44683–44690.
64 Mitra MS, Chen Z, Ren H, Harris TE, Chambers KT,
Hall AM, Nadra K, Klein S, Chrast R, Su X et al.
(2013) Mice with an adipocyte-specific lipin 1
separation-of-function allele reveal unexpected roles for
phosphatidic acid in metabolic regulation. Proc Natl
Acad Sci USA 110, 642–647.
65 Wei Y, Zou Z, Becker N, Anderson M, Sumpter R,
Xiao G, Kinch L, Koduru P, Christudass CS, Veltri
RW et al. (2013) EGFR-mediated Beclin 1
phosphorylation in autophagy suppression, tumor
progression, and tumor chemoresistance. Cell 154,
1269–1284.
66 Chang HC, Huang CC, Huang CJ, Cheng JS, Liu SI,
Tsai JY, Chang HT, Huang JK, Chou CT & Jan CR
(2008) Desipramine-induced apoptosis in human PC3
prostate cancer cells: activation of JNK kinase and
caspase-3 pathways and a protective role of [Ca2+]i
elevation. Toxicology 250, 9–14.
67 Ma J, Qiu Y, Yang L, Peng L, Xia Z, Hou LN, Fang
C, Qi H & Chen HZ (2011) Desipramine induces
apoptosis in rat glioma cells via endoplasmic reticulum
stress-dependent CHOP pathway. J Neurooncol 101,
41–48.
68 Ma J, Hou LN, Rong ZX, Liang P, Fang C, Li HF, Qi
H & Chen HZ (2013) Antidepressant desipramine leads
to C6 glioma cell autophagy: implication for the
adjuvant therapy of cancer. Anticancer Agents Med
Chem 13, 254–260.
69 Carter RE & Sorkin A (1998) Endocytosis of functional
epidermal growth factor receptor-green fluorescent
protein chimera. J Biol Chem 273, 35000–35007.
70 Morita S-Y, Ueda K & Kitagawa S (2009) Enzymatic
measurement of phosphatidic acid in cultured cells. J
Lipid Res 50, 1945–1952.
FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS
R. Shaughnessy et al.
71 Cancino J, Torrealba C, Soza A, Yuseff MI, Gravotta
D, Henklein P, Rodriguez-Boulan E & Gonzalez A
(2007) Antibody to AP1B adaptor blocks biosynthetic
and recycling routes of basolateral proteins at recycling
endosomes. Mol Biol Cell 18, 4872–4884.
FEBS Journal 281 (2014) 2172–2189 ª 2014 FEBS
Targeting endocytosis blocks oncogenic EGFR
72 Lu H, Wang X, Urvalek AM, Li T, Xie H, Yu L &
Zhao J (2014) Transformation of human ovarian
surface epithelial cells by Kruppel-like factor 8.
Oncogene 33, 10–18.
2189