Perillyl Alcohol Is an Angiogenesis Inhibitor
Heleni Loutrari, Maria Hatziapostolou, Vassoula Skouridou, Evangelia Papadimitriou,
Charis Roussos, Fragiskos N. Kolisis, and Andreas Papapetropoulos
G.P. Livanos and M. Simou Laboratories, Evangelismos Hospital, Department of Critical Care and Pulmonary Services, Medical
School, University of Athens, Athens, Greece (H.L., C.R.); Laboratory of Molecular Pharmacology, Department of Pharmacy,
University of Patras, Patras, Greece (M.H., E.P., A.P.); and Biotechnology Laboratory, School of Chemical Engineering, National
Technical University of Athens, Athens, Greece (V.S., F.K.)
Received April 23, 2004; accepted June 15, 2004
Aberrant angiogenesis is essential for the progression of solid
tumors and hematological malignancies. Thus, antiangiogenic
therapy is one of the most promising approaches to control
cancer. In the present work, we examined the ability of perillyl
alcohol (POH), a dietary monoterpene with well-established
tumor chemopreventive and chemotherapeutic activity, to in-
terfere with the process of angiogenesis. POH remarkably pre-
vented new blood vessel growth in the in vivo chicken embryo
chorioallantoic membrane assay and proved to be effective in
inhibiting the morphogenic differentiation of cultured endothe-
lial cells into capillary-like networks both in collagen gel and
Matrigel models. In addition, POH reduced the cell number in a
proliferation assay and induced apoptosis of endothelial cells
as indicated by the POH-mediated increase of caspase-3 ac-
tivity and DNA fragmentation. Consistent with the observed
antisurvival effect, POH treatment resulted in a significant inhi-
bition of Akt phosphorylation in endothelial cells. Finally, POH
was able to differentially modulate the release of two important
angiogenic regulators: vascular endothelial growth factor
(VEGF) and angiopoietin 2 (Ang2). POH decreased the release
of VEGF from cancer cells but stimulated the expression of
Ang2 by endothelial cells, indicating that it might suppress
neovascularization and induce vessel regression. Overall, these
data underscore the antiangiogenic potential of POH and sug-
gest that POH, in addition to its anticancer activity, may be an
effective agent in the treatment of angiogenesis-dependent
Perillyl alcohol (POH) (p-metha,1,7-diene-6-ol or 4-isopro-
penyl-cyclohexenecarbinol) is a naturally occurring non-nu-
tritive dietary monoterpene, found in the essential oils of
several plants (lavendin, mints, cherries, etc.) and synthe-
sized by the mevalonate pathway. It has well established
chemopreventive activity in rodent mammary, skin, liver,
lung, colon, and forestomach cancers and also chemothera-
peutic activity in pancreatic, mammary, and prostatic animal
tumor models, leading to regression of existing malignant
tumors (Gould, 1997; Belanger, 1998; Crowell, 1999). In an-
imals fed a standard chew containing 2% POH (estimated
daily intake of POH approximately 1.76 g/kg), regression of
the tumor mass has been observed (Belanger, 1998). In the
body, POH is converted to perillaldehyde, perillic acid, and
dihydroperillic acid, which exhibit biological activity. Data
from rodents have shown that tumor regression occurs when
plasma levels of perillic acid and dihydroperillic acid reach
390 to 480 ?M and 110 to 230 ?M, respectively (Belanger,
1998; Crowell, 1999). Additionally, POH has a cytostatic and
cytotoxic effect against a variety of cancer cell lines (Crowell,
1999; Burke et al., 2002; Clark et al., 2002). Treatment of
pancreatic tumor cells with POH for 2 days results in a
concentration-dependent decrease in cell proliferation, with
IC50values of 290 and 480 ?M for the human and hamster
cell lines, respectively. A similar treatment of murine B16
(F10) melanoma cells with POH inhibits cell proliferation
with an IC50of 250 ?M. Incubation of malignant hamster
pancreatic ductal epithelial cells with 100 to 500 ?M POH
caused a 2.6- to 18-fold higher rate of apoptosis and a 2-fold
higher expression of the pro-apoptotic protein Bak compared
with untreated malignant cells (Belanger, 1998, and refs.
therein). Recent evidence suggests that POH can also inhibit
cell migration in vitro (Wagner et al., 2002) and in vivo
(Teruszkin Balassiano et al., 2002), and it is capable of sen-
sitizing glioma cells to ionizing radiation and conventional
This study was supported in part by a grant from The Chios Gum Mastic
Growers Association and by funds from the Thorax Foundation.
Article, publication date, and citation information can be found at
ABBREVIATIONS: POH, perillyl alcohol; VEGF, vascular endothelial growth factor; Ang, angiopoietin; CAM, chorioallantoic membrane; BLMVEC,
bovine lung microvascular endothelial cell; HUVEC, human umbilical vein endothelial cell; PMA, phorbol 12-myristate 13-acetate; ELISA,
enzyme-linked immunosorbent assay.
THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS
Copyright © 2004 by The American Society for Pharmacology and Experimental Therapeutics
JPET 311:568–575, 2004
Vol. 311, No. 2
Printed in U.S.A.
chemotherapy (Rajesh et al., 2003). Given its proved antican-
cer activity and its favorable toxicity profile, POH is cur-
rently being tested in phase I and phase II clinical trials in
patients with refractory solid malignancies. Pharmacokinetic
studies in humans revealed that levels of 390 to 480 ?M of
the POH metabolite perillic acid are achieved in patients
receiving nontoxic doses of POH orally (Belanger, 1998;
Hudes et al., 2000; Azzoli et al., 2003).
Several mechanisms have been proposed to mediate the
antitumor effects of POH. It has been shown that POH af-
fects the expression of several regulators of cell cycle and
apoptosis (Ariazi et al., 1999; Bardon et al., 2002; Shi and
Gould, 2002). There is also evidence that POH inhibits the
post-translational isoprenylation of the Ras small GTPase
superfamily of proteins (Hohl and Lewis, 1995; Stayrook et
al., 1998; Holstein and Hohl, 2003), which are known to play
a key role in many signal transduction pathways, including
those that stimulate tumor-associated angiogenesis (Kranen-
burg et al., 2004).
The development of an angiogenic supply from the existing
vasculature is critical for the growth of solid tumors, as well
as forthe progressionof
(Saaristo et al., 2000; Bergers and Benjamin, 2003; Moehler
et al., 2003). Angiogenesis, under both physiological and dis-
ease-associated conditions, is a complex multistep process
involving the orchestrated interaction of endothelial cells
with the extracellular matrix and soluble angiogenic factors.
Among the known angiogenic factors, the families of vascular
endothelial growth factors (VEGFs) and angiopoietins (Ang1
to Ang4) have emerged as essential coordinators of angiogen-
esis, selectively targeting the endothelium through their en-
dothelial cell-specific tyrosine kinase receptors (Yancopoulos
et al., 2000; Jones et al., 2001; Ferrara et al., 2003). To date,
much effort has been directed toward discovering antiangio-
genic agents and evaluating them as cancer therapeutics
(Eskens, 2004). Among them, several dietary and nondietary
phytochemicals have been investigated for their antiangio-
genic activity in both animal models and cell culture systems
(Singh and Agarwal, 2003). In the present work, we have
investigated the potential usefulness of POH as an effective
antiangiogenic agent. We have studied the activities of POH
with regard to growth of new blood vessels in the chicken
embryo chorioallantoic membrane (CAM) in vivo, morpholog-
ical differentiation of endothelial cells in three-dimensional
collagen gel and Matrigel, endothelial cell proliferation and
apoptosis, and production of the angiogenic growth factors
VEGF and Ang2. The effect of the enzymatically synthesized
ester of POH with decanoic acid was also examined in par-
allel in most of the above-mentioned studies.
Materials and Methods
Materials. Bovine lung microvascular endothelial cells (BLM-
VECs) were obtained from Vec Technologies (Rensselear, NY), and
K562 cells were originally obtained from American Type Culture
Collection (Manassas, VA). Tissue culture plastic ware was obtained
from Corning-Costar (Corning, NY). Cell culture media and supple-
ments, fetal calf serum, trypsin, and antibiotics were obtained from
Invitrogen (Carlsbad, CA). POH was obtained from Fluka (Buchs,
Switzerland). Lipase B from Candida antarctica was kindly offered
by Novo Nordisk (Bagsvaerd, Denmark), and silica gel plates were
purchased from Merck (Darmstadt, Germany). Rat tail collagen, type
I was from BD Biosciences (San Jose, CA). The Supersignal Chemi-
luminescent Substrate was obtained from Pierce Chemical (Rock-
ford, IL). The anti-Akt and anti-phospho-Akt were from Cell Signal-
ing Technology Inc. (Beverly, MA). X-ray film was obtained from
Eastman Kodak (Rochester, NY). The anti-rabbit HRP-labeled sec-
ondary antibody was purchased from PerkinElmer Life and Analyt-
ical Sciences (Boston, MA). VEGF and Ang2 ELISA Duo set kits were
obtained from R&D Systems (Minneapolis, MN). Reagents for SDS-
polyacrylamide gel electrophoresis and western blotting were ob-
tained from Bio-Rad (Hercules, CA). The EnzChek and CyQUANT
kits for determining caspase-3 activity and cell number, respectively,
were purchased from Molecular Probes (Eugene, OR). All other re-
agents included were obtained from Sigma-Aldrich (St. Louis, MO).
Cell Culture. BLMVECs were cultured in Dulbecco’s modified
Eagle medium supplemented with 10% fetal calf serum, L-glutamine,
and antibiotics (10 U/ml penicillin and 100 mg/ml streptomycin) and
used up to passage 12. Human umbilical vein endothelial cells
(HUVECs) were isolated as previously described (Hatziapostolou et
al., 2003) and cultured in M199 supplemented with 15% fetal calf
serum, 200 ?g/ml endothelial cell growth supplement, 4 U/ml hepa-
rin sodium, L-glutamine, and antibiotics and were used at passages
1 to 5. The human lymphoblastoma K562, B16 mouse skin mela-
noma, and MDA-MB-231 human mammary gland cancer cell lines
were maintained in RPMI 1640 supplemented with 10% fetal calf
serum, L-glutamine, and antibiotics. Cells were incubated in a hu-
midified 37°C incubator containing 5% CO2.
Synthesis of POH Esters. The enzymatic synthesis and purifi-
cation of POH decanoic ester was performed as previously described
(Skouridou et al., 2003). Briefly, 5 mmol POH and 5 mmol decanoic
acid were added in 10 ml of hexane, and the esterification was
catalyzed by immobilized lipase B from C. antarctica at 50°C for 48 h
under continuous shaking. POH ester was isolated by preparative
thin-layer chromatography: the reaction mixture was concentrated
by vacuum evaporation, applied to 60 F264 silica gel plates, and
eluted with hexane/diethyl ether 8/2 (v/v). The ester was extracted
from the plates with acetone and after solvent removal was used
with no further purification. Stock solutions (1 M) of POH and POH
ester were prepared in absolute ethanol and further dilutions were
made in culture media.
CAM Assay. The in vivo chicken embryo CAM angiogenesis model
was used, as previously described (Papadimitriou et al., 2001). In
brief, Leghorn fertilized eggs were incubated for 4 days at 37°C,
when a window was opened on the egg’s shell, exposing the CAM.
The window was covered with tape, and the eggs were returned to
the incubator. Different amounts of POH (2, 10, and 20 nmol) in a
solution containing 0.1% ethanol or vehicle (0.1% ethanol) were
applied onto an area of 1 cm2(restricted by a plastic ring) of the CAM
on day 9 of embryo development. Forty-eight hours after treatment
and subsequent incubation at 37°C, CAMs were fixed in situ, excised
from the eggs, placed on slides, and left to air dry. Pictures were
taken through a stereoscope equipped with a digital camera, and the
total length of the vessels was measured using image analysis soft-
ware (Scion Image; Scion Corporation, Frederick, MD). Assays for
each test sample were carried out three times, and each experiment
included 8 to 10 eggs per data point.
In Vitro Angiogenesis Assays. Three-dimensional cultures of
BLMVECs in collagen gels were established as previously described
(Papapetropoulos et al., 1997). Briefly, rat tail collagen type I at a
final concentration of 2 mg/ml was mixed with 10? M199, neutral-
ized with sterile 1 M NaOH, and the solution held at 4°C. Cells (2 ?
106cells/ml) were added immediately to the collagen solution, and
five drops (0.1 ml each) of the mixture was put in 60-mm Petri dishes
and placed in a 37°C humidified 5% CO2incubator to permit gel
formation. Five milliliters of culture medium containing POH (0.25,
0.5, or 1.0 mM) or ethanol vehicle (0.1%) was then added over the gel
and the dishes returned to the incubator. After 2 h, phorbol 12-
myristate 13-acetate (PMA) was pipetted into the medium at a final
concentration of 100 nM, and cells were allowed to differentiate over
a 5-day period. Gels were photographed, and their diameters were
Antiangiogenic Effects of Perillyl Alcohol
monitored macroscopically. To evaluate the formation of structures
resembling tubes with lumens, cell cultures were washed with PBS
examined by phase-contrast microscopy and microphotographed.
The Matrigel tube formation assay was performed as previously
described (Pipili-Synetos et al., 1998). Briefly, Matrigel was used to
coat the wells of 96-well tissue culture plates (0.04 ml/well) and left
to solidify for 1 h at 37°C; 15,000 HUVECs were then suspended in
0.15 ml of M199 supplemented with 5% fetal calf serum and added to
each well. Different concentrations of POH (0.1, 0.5, and 1 mM) were
added to the corresponding wells simultaneously with the cells. After
6 h of incubation at 37°C, the medium was removed, the cells were
fixed, and the length of structures that resembles capillary cords was
measured in the total area of the wells using image analysis software
(Scion Image), as previously described (Papadimitriou et al., 2001;
Hatziapostolou et al., 2003).
Cell Proliferation Assay. BLMVECs or K562 cells were plated
at 2 ? 103or 15 ? 103cells/well, respectively, in a 96-well plate and
24 h later were treated with fresh media containing POH, POH ester
(0.1, 0.5, and 1 mM), or ethanol vehicle (0.1%) and further cultured
for 48 to 72 h. Thereafter, the cell number was measured by using the
CyQUANT cell proliferation assay kit according to the manufactur-
er’s instructions. This assay has a linear detection range extending
from 50 to 50,000 cells in 200-?l volumes using the dye provided by
the kit at 1? concentration. Briefly, medium was removed from the
wells, cells were carefully washed once with PBS, and then 200 ?l of
CyQUANT GR dye/cell lysis buffer was added to each well and
incubated for 5 min at room temperature, protected from light. The
sample fluorescence was measured in a fluorescence microplate
reader using excitation and emission filters at 480 and 520 nm,
respectively. A reference standard curve using BLMVECs or K562
cells (from 50–50,000) was created for converting sample fluores-
cence to cell numbers.
Apoptosis Assays. Cell apoptosis was monitored by measuring
caspase-3 activity and by determining DNA fragmentation. For the
caspase-3 activity assay, K562 cells (1 ? 106) or BLMVECs grown to
confluence in 12-well plates were incubated with fresh media con-
taining various concentrations of POH (0.1, 0.5, and 1 mM), POH
ester (1 mM), ethanol vehicle (0.1%), or 5 to 10 ?M cyclohexamide as
a positive apoptosis control. After 16, 24, 48, or 72 h, cells floating in
the supernatant were combined with the adherent cells that were
collected by trypsinization, and cell pellets were washed once with
PBS. Samples were then processed according to the EnzCheck
Caspase-3 Assay kit instructions. Briefly, cells were lysed with 60 ?l
of lysis buffer, lysates were collected by centrifugation at 13,000g for
20 min at 4°C, and protein concentrations were measured. Fifty
microliters of lysates were transferred to 96-well microplates and
incubated for 30 min with a caspase-3 substrate. The release of the
fluorescent product was measured using a fluorescence microplate
reader (excitation at 350 nm and emission detection at 450 nm).
For the DNA fragmentation assay, confluent cultures of cells
grown in 100-mm dishes were incubated with fresh media containing
1 mM POH, 0.1% ethanol vehicle, or 5 ?M cyclohexamide for 48 h.
Floating and adherent cells were then harvested, washed once with
PBS, and cytosolic DNA was prepared from cell pellets as described
by (Leist et al., 1997) with some modifications. Briefly, cell pellets
were lysed in 100 ?l of lysis buffer (20 mM Tris-HCl, 10 mM EDTA,
and 1% Triton X-100, pH 8.0), the cytosolic fraction was collected by
centrifugation at 13,000g for 20 min at 4°C, and protein concentra-
tion was determined. Cytosol aliquots containing equal amounts of
protein were extracted with phenol/chloroform. One-tenth volume of
3 M sodium acetate was added to the solution, and DNA was pre-
cipitated by adding an equal volume of isopropanol. After storing at
?20°C overnight, a DNA pellet was obtained by centrifugation at
13,000g for 15 min at 4°C and washed once with 75% ethanol. The
pellet was dried and resuspended in 20 ?l of 20 mM Tris-HCl, pH 8.0.
After digesting RNA with RNase (0.1 mg/ml) at 37°C for 30 min,
samples (10 ?l) were electrophoresed through a 1% agarose gel, and
DNA was photographed under visualization with UV light.
Western Blotting. BLMVECs grown to confluence in six-well
plates were treated with 0.5 mM POH or 0.1% ethanol for the
indicated time periods (15, 30, 60, and 120 min) and then subjected
to cell lysis and western blotting. Cells were collected and solubilized
with ice-cold cell lysis buffer (50 mM Tris-HCl, pH 7.6, 150 mM NaCl,
50 mM NaF, 1 mM Na3VO4, 0.5% sodium deoxycholate, 1 mM EDTA,
0.1 mM EGTA, 1% Triton-X, 1% SDS, 1 mM PMSF, 10 ?g/ml apro-
tinin, 5 ?g/ml leupeptin, and 10 ng/ml pepstatin). After centrifuga-
tion at 13,000g for 20 min at 4°C, protein concentration of superna-
tants was determined. Lysates containing equal amounts of protein
were resolved by SDS-polyacrylamide gel electrophoresis on 10%
gels, transferred to nitrocellulose membranes, and probed with pri-
mary antibodies that specifically recognize the Ser473 phosphory-
lated form of Akt or total Akt. Immunoreactive bands were visual-
ized by enhanced chemiluminescence detection. A similar protocol
was followed for K562 cells; primary antibodies recognizing Erk1/2
when phosphorylated at Thr202/Tyr204 or total levels of endogenous
Erk1/2 were used.
ELISA for VEGF and Ang2 Detection. Cells grown to conflu-
ence in 24-well plates were treated with POH (0.1, 0.2, 0.5, and 1
mM), POH ester (1 mM), or vehicle (0.1% ethanol) over various time
periods (24–72 h). Conditioned media from these cultures were an-
alyzed by a VEGF or Ang2 ELISA according to the manufacturer’s
specifications. In brief, 100 ?l of sample dilution in assay diluent (1%
bovine serum albumin in PBS) were added to a 96-well microplate
precoated with polyclonal antibody specific for human VEGF or
human Ang2. Recombinant human VEGF or Ang2 was used for the
standard curve. After incubation for 2 h at room temperature, the
wells were washed, and polyclonal biotinylated anti-human VEGF or
Ang2 antibodies were added. Incubation was continued as above,
plates were washed, and streptavidin conjugated to horseradish
peroxidase was added and further incubated for 20 min. Plates were
washed again, developed using the substrate solution (equal volumes
of H2O2and tetramethylbenzidine), and the reaction was stopped by
adding 2 N H2SO4. The optical density was measured at 450 nm with
correction wavelength set at 630 nm. The standard curve was gen-
erated using the GraphPad Prism 4 (GraphPad Software Inc., San
Diego, CA) with a four-parameter logistic curve fit.
Data Analysis and Statistics. Data are presented as means of
the percentage of control ? S.E.M. of the indicated number of obser-
vations. Statistical comparisons between groups were performed us-
ing the one-way ANOVA followed by a post hoc test (Neuman-Keuls,
Dunnett’s, or least significant difference), as appropriate. Differ-
ences among means were considered significant when p ? 0.05.
POH Suppresses Angiogenesis in the CAM. To evalu-
ate the effects of POH on neovascularization in vivo, we used
the CAM model of angiogenesis. The results from the mor-
phometric analysis (Fig. 1A) showed that POH decreased the
number of CAM vessels in a dose-dependent manner up to
23 ? 1% for the 20 nmol dose. For comparison, it is noted that
dexamethasone, a steroid known to inhibit angiogenesis, in-
hibited vessel length in the same assay by 10.5 and 12% at 8
and 80 nmol, respectively. Figure 1B shows representative
pictures of the CAM vascular network following vehicle or
POH treatment. The decrease in the number of CAM vessels
was not due to toxicity, as verified on CAM paraffin sections
stained with eosin-hematoxylin (H. Loutrari and A. Papa-
petropoulos, data not shown).
POH Attenuates Capillary-Like Organization of En-
dothelial Cells in Vitro. The effect of POH on the differen-
tiation of endothelial cells into tube-like structures was stud-
ied in two different in vitro angiogenesis models, namely the
collagen gel and the Matrigel assay. In the first model, three-
Loutrari et al.
dimensional cultures of BLMVECs were preincubated with
POH or vehicle, and vessel-like structure formation was
stimulated by addition of PMA. As early as 24 h after PMA
treatment, control cells formed networks that became more
complex with prolonged incubation (Fig. 2A) and caused a
macroscopically visible contraction of the collagen gel droplet
(Fig. 2B, left). In contrast, pretreatment with POH induced a
concentration-dependent inhibition of cell organization into
networks, the cells remained interspersed throughout the gel
for the entire incubation period (Fig. 2A), and collagen gel
contraction was prevented (Fig. 2B, right).
Similar results were obtained with the Matrigel in vitro
angiogenesis assay. Culturing untreated HUVECs on Matri-
gel triggered their morphological differentiation in struc-
tures imitating geometric tubule-like networks (Fig. 3B). In
contrast, as revealed by microscopic observation (Fig. 3B)
and morphometric analysis (Fig. 3A), treatment with POH
caused a concentration-dependent reduction of the total
length of the above-mentioned structures.
POH Reduces Endothelial Cell Number and Inhibits
Cell Survival. Signal transduction pathways promoting en-
dothelial cell proliferation and survival are essential for the
angiogenic process. We observed (Fig. 4) that treatment of
exponentially growing endothelial cells with POH caused a
concentration-dependent decrease in cell number in a cell
proliferation assay. In contrast, addition of the decanoic acid
ester of POH was without effect. Furthermore, in line with
earlier studies on a variety of mammalian cancer cells, we
found that POH induced a concentration-dependent reduc-
tion in cell number of K562 cells. Incubation with 0.1, 0.5,
and 1.0 mM POH for 72 h reduced K562 cell number to
72.1 ? 5.0, 51.5 ? 4.1, and 32.4 ? 2.0% of control.
We also examined the potential effect of POH on apoptosis
induction in BLMVECs by evaluating two different apoptosis
Fig. 2. POH inhibits differentiation of endothe-
lial cells in the collagen gel assay. BLMVECs
cultured in collagen gels (2 ? 106cells/ml) were
treated with 0.5 mM POH or 0.1% ethanol (con-
trol). Two hours later, PMA (100 nM) was added,
and cells were further maintained in culture for
5 days. A, phase-contrast photomicrographs of
control or POH-treated cells. B, macroscopic pho-
tomicrograph of collagen gels from control (left)
and POH-treated (right) cultures.
Fig. 1. Effect of POH on the CAM. POH or vehi-
cle was applied onto 1 cm2of the CAM on day 9
and incubated for 48 h at 37°C. CAMs were fixed
and excised from the eggs. A, total length of the
vessel network was measured using image anal-
ysis software. Results are expressed as means ?
S.E.M.; n ? 24; ???, p ? 0.001 from the control.
B, representative photographs showing the CAM
vascular network following treatment with vehi-
cle (control) or with 20 nmol POH.
Antiangiogenic Effects of Perillyl Alcohol
indicators: caspase-3 activation and DNA fragmentation. As
shown in Fig. 5A, treatment of cells with 1 mM POH for 48 h
caused a statistically significant increase in caspase-3 activ-
ity (246.5 ? 34.7% of control); when cells were treated with 5
?M cycloheximide, a much greater increase in caspase activ-
ity was observed (1118.3 ? 46.1% of control). However, incu-
bation of endothelial cells with lower POH concentrations or
with the decanoic ester of POH had no effect. It has been
previously reported that POH induces apoptosis in several
cancer cell lines. In agreement with these observations, POH
caused a concentration-dependent stimulation of caspase-3
activity in K562 cells: after incubation with 0.1, 0.2, and 0.5
mM POH for 72 h, caspase-3 activity was 162.9 ? 7.0,
336.5 ? 30.0, and 423.7 ? 49.0% of control, respectively,
whereas cyclohexamide (10 ??) caused a 264.8 ? 6% in-
In addition, POH-treated BLMVECs exhibited a character-
istic cytosolic DNA laddering pattern, very similar to that
observed in cells incubated with cyclohexamide, as a result of
nuclear DNA fragmentation of apoptotic cells (Fig. 5B).
POH Reduces Akt Activation in Endothelial Cells. In
an attempt to explore the signaling mechanism mediating
the apoptotic activity of POH in BLMVEC, we treated the
cells with POH (0.5–1.0 mM) for various incubation times
(15–120 min) and determined the levels of phosphorylated
Akt relative to total Akt. As illustrated in Fig. 6, incubation
with POH caused a reduction of Akt phosphorylation that
was sustained over the examined incubation periods without
affecting total Akt levels. In contrast, POH did not alter Akt
phosphorylation levels in K562 lymphoblastoma cells (H.
Loutrari and A. Papapetropoulos, data not shown), although
it inhibited cell proliferation and induced apoptosis in this
cell line. Interestingly, we observed that POH inhibited
talERK1/2 ratios were 0.46 and 0.2 for vehicle- and POH-
treated cultures after 15 min of incubation).
Differential Modulation of VEGF and Ang2 Expres-
sion by POH. The effect of POH or POH ester on VEGF and
Ang2 release from cancer and endothelial cells, respectively,
was examined by ELISA. As illustrated in Fig. 7, treatment
of K562 human lymphoblastoma cells with POH caused a
concentration- and time-dependent decrease in the basal pro-
duction of VEGF. A similar response to POH treatment,
although to a lesser degree, was noted with the B16 mouse
skin melanoma and the MDA-MB-231 human mammary
gland cancer cell lines. Following incubation with 0.2, 0.5,
and 1.0 mM POH for 24 h, VEGF release was 95.4 ? 0.4,
88.1 ? 1.5, and 65.9 ? 1.3% of control, respectively for B16
cells and 89.6 ? 2.7, 81.5 ? 2.6, and 69.3 ? 3.1% of control,
respectively for MDA-MB-231 cells.
In contrast, POH treatment induced a concentration-de-
pendent increase in Ang2 release from BLMVECs (Fig. 8). It
should be noted that the decanoic ester of POH had only
Fig. 3. POH inhibits differentiation of endothe-
lial cells in the Matrigel assay. HUVECs (105
cells/ml) were added onto Matrigel-coated wells
in 96-well plates in the presence of POH or ve-
hicle (control) and incubated for 6 h. A, length of
the tube network was measured in the total well
area. Results are expressed as means ? S.E.M.;
n ? 9; ??, p ? 0.01 from control. B, representa-
tive photomicrographs showing the formation of
tube-like structures on Matrigel after control or
POH (1 mM) treatment.
Fig. 4. Effect of POH on endothelial cell proliferation. BLMVECs were
plated in a 96-well plate at 2000 cells/well. The next day, POH, POH ester
(E), or vehicle were added, cultures were incubated for 48 h, and cell
numbers were measured using the CyQUANT cell proliferation assay kit.
Results are expressed as means ? S.E.M.; n ? 9; ?, p ? 0.05 from control.
Loutrari et al.
minimal effects on the secretion of both angiogenic factors
(Figs. 7 and 8).
The monoterpene POH has been previously shown to exert
a significant antitumor activity in animal models, which has
been attributed to the cytostatic and cytotoxic actions of POH
on cancer cells (Gould, 1997; Belanger, 1998; Crowell, 1999).
However, in vivo antitumor activity could also result from
inhibition of neovascularization. To evaluate the ability of
POH to interfere with the angiogenic process in vivo, we used
a well-established model of neo-vessel formation and ob-
served that POH dose-dependently reduced the number of
vessels in the CAM. Since neovascularization is a multistep
process that requires coordinated activation of many signal-
ing pathways triggered by a number of angiogenic inducers
(Yancopoulos et al., 2000; Jones et al., 2001), we speculated
that POH might affect different stages of angiogenesis. To
test this hypothesis, we designed a series of in vitro experi-
ments evaluating the potentially pleiotropic activities of
POH. Based on preliminary experiments and on information
in the literature regarding the action of POH on tumor cell
lines, 0.1 to 1 mM was used in all our assays (Belanger, 1998,
and refs. therein).
The effect of POH on the differentiation of endothelial cells
was examined in two different in vitro angiogenesis models.
It is well established that culture of endothelial cells in a
three-dimensional scaffold of extracellular matrix protein ac-
celerates their morphological differentiation into tube-like
structures. This process involves dramatic changes in endo-
thelial cell cytoskeletal dynamics, resulting in enhanced mo-
tility and alterations in cell shape (Connolly et al., 2002;
Davis et al., 2002). We observed that POH inhibited in a
concentration-dependent manner the organization of endo-
thelial cells into structures that resemble capillaries in both
collagen gel and Matrigel assays. Furthermore, POH pre-
vented the contraction of collagen gels, suggesting that it
might reduce endothelial cell actin stress fiber formation
(Hoang et al., 2004). It has been widely demonstrated that
Rho GTPases interact with the cytoskeleton and are, thus,
major regulators of cytokinesis (Bishop and Hall, 2000). More
recent evidence suggests that Rho GTPases participate in
signaling pathways that control endothelial cell phenotype
during neovascularization (Connolly et al., 2002; Cascone et
al., 2003; Hoang et al., 2004). Thus, it is possible that POH
attenuates the formation of tube-like networks by inhibiting
the post-translational isoprenylation and association with
the membrane and subsequent activation of Rho GTPases in
endothelial cells through a mechanism that might be similar
to the one observed in tumor cells (Hohl and Lewis, 1995;
Stayrook et al., 1998; Holstein and Hohl, 2003).
Endothelial cell proliferation and survival are important
steps in the angiogenic process. Most pro-angiogenic growth
factors stimulate endothelial cell proliferation (Yancopoulos
et al., 2000; Jones et al., 2001; Ferrara et al., 2003), whereas
Fig. 5. POH induces apoptosis in endothelial cells.
A, BLMVECs grown to confluence were treated
with POH, POH ester (E), or vehicle (control) for
48 h and then analyzed for caspase-3 activity. Re-
sults are expressed as means ? S.E.M.; n ? 9; ?,
p ? 0.05 from control. Caspase-3 activity in cyclo-
heximide-treated cells was 1118.3 ? 46.1% of con-
trol. B, confluent cultures of BLMVECs were incu-
bated with 5 ?? cyclohexamide (CX), 1 mM POH,
or vehicle (C) for 48 h. Cells were then lysed, and
DNA was extracted from cytosolic aliquots con-
taining equal amounts of protein. DNA samples
and markers (M) were electrophoresed through a
1% agarose gel and photographed under UV light.
Control samples contain no cytosolic DNA since
vehicle-treated cultures do not undergo pro-
grammed cell death. Similar results were obtained
in two experiments.
Fig. 6. POH inhibits Akt phosphorylation in endothelial cells. BLMVECs
grown to confluence were incubated with 0.5 mM POH (P) or vehicle (C)
for the indicated times (minutes) and then solubilized with lysis buffer.
Equal amounts of protein lysates were run on a 10% SDS-PAGE, trans-
ferred to a nitrocellulose membrane, and probed with phospho-specific
anti-Akt or anti-total Akt antibodies. Immunoreactive bands were visu-
alized with ECL. The pAkt/Akt ratios for POH-treated cells were 39, 85,
57, and 37% of vehicle-treated cells (control) after 15, 30, 60, and 120 min.
Fig. 7. POH reduces VEGF release from cancer cells. K562 cells (3 ? 106
cells/ml) were treated with POH, POH ester (E), or vehicle (control). At
the indicated times, culture supernatants were collected and measured
by ELISA for the presence of VEGF. Results are expressed as means ?
S.E.M.; n ? 9; ?, p ? 0.05 from control. Concentrations of POH and E are
Antiangiogenic Effects of Perillyl Alcohol
other important molecules involved in neovessel formation,
like Ang1, do not directly stimulate endothelial cell growth
but rather stabilize vascular networks by promoting endo-
thelial cell survival (Papapetropoulos et al., 1999, 2000).
POH has well-established cytostatic and apoptotic activities
in a variety of mammalian cancer cells (Crowell, 1999; Burke
et al., 2002; Clark et al., 2002). We also observed the induc-
tion of programmed cell death in the cancer cell lines used in
our study. Interestingly, POH also reduced endothelial cell
number and stimulated apoptosis of endothelial cells, as in-
dicated by the increase in caspase-3 activity and DNA frag-
mentation. The reduction in EC number brought about by
low POH concentrations might be due to inhibition of cell
proliferation, whereas the lower cell number seen with in-
creased POH concentrations is likely due to stimulation of
apoptosis. To determine pathways potentially involved in
POH’s effect on apoptosis, we tested the ability of POH to
affect activation of Akt, a key kinase in survival signaling
pathways (Franke et al., 2003). We found that POH de-
creased the levels of phosphorylated (active) form of Akt,
indicating that POH-induced suppression of endothelial cell
survival could be mediated through inhibition of Akt signal-
ing. An interesting observation was that although POH very
efficiently attenuated cell proliferation and induced apopto-
sis in K562 cancer cells, it had no effect on Akt phosphory-
lation in these cells (H. Loutrari and A. Papapetropoulos,
unpublished data). In line with the notion that POH blocks
small GTPase signaling, POH-mediated apoptosis in K562
cells correlated with an inhibition of ERK1/2 phosphoryla-
tion, a downstream target of the Ras/Raf pathway. These
results suggest that the pro-apoptotic action of POH may
involve different molecular targets in endothelial and cancer
We finally determined the effects of POH on angiogenic
growth factor production. Due to their proven importance in
neovascularization, we focused on two such factors, namely
VEGF and Ang2. VEGF is mainly produced by cancer cells, in
response to multiple stimuli (activation of oncogenic Ras
proteins, hypoxia, and ultraviolet radiation) and activates its
cognate receptors on endothelial cells leading to prolifera-
tion, migration, survival, and vascular permeability (Ferrara
et al., 2003). Ang2, on the other hand, is primarily produced
by endothelial cells and exerts its biological actions through
the antagonism of Ang1-stimulated phosphorylation of Tie2
on vascular endothelium (Yancopoulos et al., 2000; Jones et
al., 2001). In the absence of VEGF, Ang2 stimulates vessel
regression, whereas in the presence of VEGF, neovascular-
ization is favored (Holash et al., 1999; Yancopoulos et al.,
2000; Vajkoczy et al., 2002). In our experiments, we observed
that POH down-regulated the basal production of VEGF in
three different cancer cell lines and up-regulated the release
of Ang2 from endothelial cells, favoring the balance of the two
growth factors toward a state that leads to vessel regression.
In most of our experiments, we examined the effect of POH
ester with decanoic acid in parallel with POH. In a previous
study using the flavonoid rutin, we noted that addition of a
fatty acid chain on this compound improved its biological
activity that was probably due to an increase of its lipophi-
licity (Kodelia et al., 1994). However, in the present work, we
observed that esterification of POH on its unique ?OH group
with decanoic acid essentially abolished its biological activity
suggesting that a free ?OH group is indispensable for max-
imal POH biological activity. Further experiments using al-
ternative POH esters with a smaller carbon chain length are
currently in progress.
In conclusion, we have provided evidence that POH pos-
sesses antiangiogenic properties (inhibition of new vessel
growth in the CAM assay, inhibition of endothelial cell pro-
liferation and organization into tube-like structures, alter-
ation in the production of angiogenic growth factors, and
induction of apoptosis) that complement its toxic effects on
tumor cells. Thus, the combined effects of POH on tumor and
endothelial cells might explain its strong anticancer activity.
The observed inhibition of VEGF release coupled to the in-
crease in Ang2 production provides an additional mechanism
through which POH treatment might cause tumor regression
by destroying the vascular networks supporting the tumor
Ariazi EA, Satomi Y, Ellis MJ, Haag JD, Shi W, Sattler CA, and Gould MN (1999)
Activation of the transforming growth factor beta signaling pathway and induction
of cytostasis and apoptosis in mammary carcinomas treated with the anticancer
agent perillyl alcohol. Cancer Res 59:1917–1928.
Azzoli CG, Miller VA, Ng KK, Krug LM, Spriggs DR, Tong WP, Riedel ER, and Kris
MG (2003) A phase I trial of perillyl alcohol in patients with advanced solid
tumors. Cancer Chemother Pharmacol 51:493–498.
Bardon S, Foussard V, Fournel S, and Loubat A (2002) Monoterpenes inhibit prolif-
eration of human colon cancer cells by modulating cell cycle-related protein ex-
pression. Cancer Lett 181:187–194.
Belanger JT (1998) Perillyl alcohol: applications in oncology. Altern Med Rev 3:448–
Bergers G and Benjamin LE (2003) Tumorigenesis and the angiogenic switch. Nat
Rev Cancer 3:401–410.
Bishop AL and Hall A (2000) Rho GTPases and their effector proteins. Biochem J 348
Burke YD, Ayoubi AS, Werner SR, McFarland BC, Heilman DK, Ruggeri BA, and
Crowell PL (2002) Effects of the isoprenoids perillyl alcohol and farnesol on
apoptosis biomarkers in pancreatic cancer chemoprevention. Anticancer Res 22:
Cascone I, Giraudo E, Caccavari F, Napione L, Bertotti E, Collard JG, Serini G, and
Bussolino F (2003) Temporal and spatial modulation of Rho GTPases during in
vitro formation of capillary vascular network. Adherens junctions and myosin light
chain as targets of Rac1 and RhoA. J Biol Chem 278:50702–50713.
Clark SS, Perman SM, Sahin MB, Jenkins GJ, and Elegbede JA (2002) Antileukemia
activity of perillyl alcohol (POH): uncoupling apoptosis from G0/G1 arrest suggests
that the primary effect of POH on Bcr/Abl-transformed cells is to induce growth
arrest. Leukemia 16:213–222.
Connolly JO, Simpson N, Hewlett L, and Hall A (2002) Rac regulates endothelial
morphogenesis and capillary assembly. Mol Biol Cell 13:2474–2485.
Crowell PL (1999) Prevention and therapy of cancer by dietary monoterpenes. J Nutr
Fig. 8. POH stimulates Ang2 release from endothelial cells. BLMVECs
grown to confluence were incubated with POH, POH ester (E), or vehicle
(control). At the indicated times, culture supernatants were collected and
measured by ELISA for the presence of Ang2. Results are expressed as
means ? S.E.M.; n ? 9; ?, p ? 0.05 from control. Concentrations of POH
and E are in millimolar.
Loutrari et al.
Davis GE, Bayless KJ, and Mavila A (2002) Molecular basis of endothelial cell Download full-text
morphogenesis in three-dimensional extracellular matrices. Anat Rec 268:252–
Eskens FA (2004) Angiogenesis inhibitors in clinical development: where are we now
and where are we going? Br J Cancer 90:1–7.
Ferrara N, Gerber HP, and LeCouter J (2003) The biology of VEGF and its receptors.
Nat Med 9:669–676.
Franke TF, Hornik CP, Segev L, Shostak GA, and Sugimoto C (2003) PI3K/Akt and
apoptosis: size matters. Oncogene 22:8983–8998.
Gould MN (1997) Cancer chemoprevention and therapy by monoterpenes. Environ
Health Perspect 105(Suppl 4):977–979.
Hatziapostolou M, Katsoris P, and Papadimitriou E (2003) Different inhibitors of
plasmin differentially affect angiostatin production and angiogenesis. Eur J Phar-
Hoang MV, Whelan MC, and Senger DR (2004) Rho activity critically and selectively
regulates endothelial cell organization during angiogenesis. Proc Natl Acad Sci
Hohl RJ and Lewis K (1995) Differential effects of monoterpenes and lovastatin on
RAS processing. J Biol Chem 270:17508–17512.
Holash J, Wiegand SJ, and Yancopoulos GD (1999) New model of tumor angiogen-
esis: dynamic balance between vessel regression and growth mediated by angio-
poietins and VEGF. Oncogene 18:5356–5362.
Holstein SA and Hohl RJ (2003) Monoterpene regulation of Ras and Ras-related
protein expression. J Lipid Res 44:1209–1215.
Hudes GR, Szarka CE, Adams A, Ranganathan S, McCauley RA, Weiner LM, Langer
CJ, Litwin S, Yeslow G, Halberr T, et al. (2000) Phase I pharmacokinetic trial of
perillyl alcohol (NSC 641066) in patients with refractory solid malignancies. Clin
Cancer Res 6:3071–3080.
Jones N, Iljin K, Dumont DJ, and Alitalo K (2001) Tie receptors: new modulators of
angiogenic and lymphangiogenic responses. Nat Rev Mol Cell Biol 2:257–267.
Kodelia G, Athanasiou K, and Kolisis FN (1994) Enzymatic synthesis of butyryl-
rutin ester in organic solvents and its cytogenetic effects in mammalian cells in
culture. Appl Biochem Biotechnol 44:205–212.
Kranenburg O, Gebbink MF, and Voest EE (2004) Stimulation of angiogenesis by
Ras proteins. Biochim Biophys Acta 1654:23–37.
Leist M, Volbracht C, Kuhnle S, Fava E, Ferrando-May E, and Nicotera P (1997)
Caspase-mediated apoptosis in neuronal excitotoxicity triggered by nitric oxide.
Mol Med 3:750–764.
Moehler TM, Ho AD, Goldschmidt H, and Barlogie B (2003) Angiogenesis in hema-
tologic malignancies. Crit Rev Oncol Hematol 45:227–244.
Papadimitriou E, Polykratis A, Courty J, Koolwijk P, Heroult M, and Katsoris P
(2001) HARP induces angiogenesis in vivo and in vitro: implication of N or C
terminal peptides. Biochem Biophys Res Commun 282:306–313.
Papapetropoulos A, Desai KM, Rudic RD, Mayer B, Zhang R, Ruiz-Torres MP,
Garcia-Cardena G, Madri JA, and Sessa WC (1997) Nitric oxide synthase inhibi-
tors attenuate transforming-growth-factor-beta 1-stimulated capillary organiza-
tion in vitro. Am J Pathol 150:1835–1844.
Papapetropoulos A, Fulton D, Mahboubi K, Kalb RG, O’Connor DS, Li F, Altieri DC,
and Sessa WC (2000) Angiopoietin-1 inhibits endothelial cell apoptosis via the
Akt/surviving pathway. J Biol Chem 275:9102–9105.
Papapetropoulos A, Garcia-Cardena G, Dengler TJ, Maisonpierre PC, Yancopoulos
GD, and Sessa WC (1999) Direct actions of angiopoietin-1 on human endothelium:
evidence for network stabilization, cell survival and interaction with other angio-
genic growth factors. Lab Investig 79:213–223.
Pipili-Synetos E, Papadimitriou E, and Maragoudakis ME (1998) Evidence that
platelets promote tube formation by endothelial cells in vitro. Br J Pharmacol
Rajesh D, Stenzel RA, and Howard SP (2003) Perillyl alcohol as a radio-/
chemosensitizer in malignant glioma. J Biol Chem 278:35968–35978.
Saaristo A, Karpanen T, and Alitalo K (2000) Mechanisms of angiogenesis and their
use in the inhibition of tumor growth and metastasis. Oncogene 19:6122–6129.
Shi W and Gould MN (2002) Induction of cytostasis in mammary carcinoma cells
treated with the anticancer agent perillyl alcohol. Carcinogenesis 23:131–142.
Singh RP and Agarwal R (2003) Tumor angiogenesis: a potential target in cancer
control by phytochemicals. Curr Cancer Drug Targets 3:205–217.
Skouridou V, Stamatis H, and Kolisis FN (2003) Use of essential oils as media for the
enatioselective esterification of the monoterpene perillyl alcohol catalysed by
lipase. Eur J Lipid Sci Tech 105:115–120.
Stayrook KR, McKinzie JH, Barbhaiya LH, and Crowell PL (1998) Effects of the
antitumor agent perillyl alcohol on H-Ras vs. K-Ras farnesylation and signal
transduction in pancreatic cells. Anticancer Res 18:823–828.
Teruszkin Balassiano I, Alves de Paulo S, Henriques Silva N, Curie Cabral M,
Gibaldi D, Bozza M, Orlando da Fonseca C, and Da Gloria da Costa Carvalho M
(2002) Effects of perillyl alcohol in glial C6 cell line in vitro and anti-metastatic
activity in chorioallantoic membrane model. Int J Mol Med 10:785–788.
Vajkoczy P, Farhadi M, Gaumann A, Heidenreich R, Erber R, Wunder A, Tonn JC,
Menger MD, and Breier G (2002) Microtumor growth initiates angiogenic sprout-
ing with simultaneous expression of VEGF, VEGF receptor-2 and angiopoietin-2.
J Clin Investig 109:777–785.
Wagner JE, Huff JL, Rust WL, Kingsley K, and Plopper GE (2002) Perillyl alcohol
inhibits breast cell migration without affecting cell adhesion. J Biomed Biotechnol
Yancopoulos GD, Davis S, Gale NW, Rudge JS, Wiegand SJ, and Holash J (2000)
Vascular-specific growth factors and blood vessel formation. Nature (Lond) 407:
Address correspondence to: Dr. Andreas Papapetropoulos, Laboratory for
Molecular Pharmacology, Department of Pharmacy, University of Patras,
Patras, Greece 26504. E-mail: firstname.lastname@example.org
Antiangiogenic Effects of Perillyl Alcohol