Mol. Nutr. Food Res. 2010, 54, 551–558 DOI 10.1002/mnfr.

200900012 551

RESEARCH ARTICLE

b-Sitosterol exhibits anti-inflammatory activity in human

aortic endothelial cells

Stella Loizou1, Ioannis Lekakis2, George P. Chrousos3 and Paraskevi Moutsatsou1

1
Department of Biological Chemistry, Medical School, University of Athens, Athens, Greece
2
Second Department of Cardiology, University General Hospital ‘‘Attikon’’, Athens, Greece
3
First Department of Pediatrics and Unit on Endocrinology, Metabolism and Diabetes, University of Athens,
Athens, Greece

b-Sitosterol, normally present in vegetable-containing diets, comprises an important compo- Received: January 9, 2009
nent of cholesterol controlling functional foods. It has been associated with cardiovascular Revised: May 19, 2009
protection, exerting its effect mainly through increasing the antioxidant defense system and Accepted: June 12, 2009
effectively lowering the serum cholesterol levels in humans. However, its anti-inflammatory
effect on endothelium is unknown. Attachment of leukocytes to the vascular endothelium and
the subsequent migration of cells into the vessel wall are early events in atherogenesis, this
process requiring the expression of endothelial adhesion molecules. We examined the effect of
b-sitosterol (0.1–200 mM) on (i) the expression of vascular adhesion molecule 1 and intracellular
adhesion molecule 1 by cell ELISA and (ii) the attachment of monocytes (U937 cells) in tumor
necrosis factor-a (TNF-a)-stimulated human aortic endothelial cells (HAECs) by adhesion
assay. The effect on nuclear factor-kB phosphorylation was also examined via a cell-based
ELISA kit. Results showed that b-sitosterol inhibits significantly vascular adhesion molecule 1
and intracellular adhesion molecule 1 expression in TNF-a-stimulated HAEC as well as the
binding of U937 cells to TNF-a-stimulated HAEC and attenuates the phosphorylation of
nuclear factor-kB p65. This study extends existing data regarding the cardioprotective effect of
b-sitosterol and provides new insights into understanding the molecular mechanism under-
lying the beneficial effect of b-sitosterol on endothelial function.

Keywords:
Adhesion molecules / Atherosclerosis / b-Sitosterol / Human aortic endothelial cell /
Nuclear factor-kB

1 Introduction compounds are not synthesized in mammals and are there-

fore derived solely from the diet, the most common being
Plant sterols (phytosterols) are naturally occurring plant b-sitosterol, campesterol and stigmasterol. It is well estab-
constituents that are structurally similar to cholesterol. These lished that dietary plant sterols lower plasma cholesterol
concentrations by inhibiting intestinal cholesterol absorption
Correspondence: Dr. P. Moutsatsou, Department of Biological [1]. Consumption of 1.8–2.0 g/day of plant sterols has been
Chemistry, Medical School, University of Athens, 75 Mikras shown to lower both total and LDL cholesterol concentrations
Asias Street, Goudi, Athens 11527, Greece
by 10–15% in a variety of different population groups [2, 3].
E-mail: [email protected]
Fax:130-210 7462682 Because it has been demonstrated that plant sterols have
additive effects in lowering LDL cholesterol when combined
Abbreviations: BCECF-AM, 20 ,70 -bis-(2-cabroxyethyl)-5-(and-6)- with statins [4], the National Cholesterol Education Program
carboxy-fluorescein acetoxymethyl ester; CAM, cell adhesion Adult Treatment Panel III has recognized the use of
molecule; FBS, fetal bovine serum; HAEC, human aortic
phytosterols as a key element of dietary therapy [5]. This has
endothelial cells; ICAM-1, intracellular adhesion molecule 1;
led to the development of functional foods containing high
MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide; NFkB, nuclear factor-kB; PR, phenol red; RT, room contents of these plant molecules or their esters as cholesterol
temperature; TNF-a, tumor necrosis factor-a; VCAM-1, vascular controlling foods. In addition, anti-atherogenic effects of
adhesion molecule 1 dietary plant sterols have been associated with inhibition of

& 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mnf-journal.com
552 S. Loizou et al. Mol. Nutr. Food Res. 2010, 54, 551–558

proinflammatory cytokine production in ApoE-KO mice [6]. It 0.01% solution (Gibco BRL, Invitrogen). Cell viability was
is remarkable that b-sitosterol, a natural compound normally estimated by a modification of the 3-(4,5-dimethylthiazol-2-
present in vegetable-containing diets, shows beneficial effects yl)-2,5-diphenyltetrazolium bromide (MTT) assay [13],
directly on endothelial and monocytic cells [7]. Furthermore, which determines the metabolically active mitochondria of
in vivo studies support the topical anti-inflammatory effect of cells. Briefly, cells were plated in their growth medium at a
b-sitosterol [8, 9], while in vitro studies suggest the involve- density of 6000 cells/well in 96 flat-bottomed well plates.
ment of b-sitosterol in the arachidonic acid cascade [10]. Twenty-four hours after plating, b-sitosterol from soybean
Atherosclerosis is considered a chronic inflammatory (purityZ97%), (S9889, Sigma-Aldrich, Germany) was added
process with increased oxidative stress in which the adhe- at final concentrations ranging from 0.1 to 100 mM in
sion of monocytes to the vascular endothelium and their DMEM phenol red (PR)-free medium (Gibco BRL Invitro-
subsequent migration into the vessel wall are the pivotal gen). Cells in their growth medium – without test
early events in atherogenesis [11]. The interaction between substances – were used as control samples (vehicle). After
monocytes and vascular endothelial cells is mediated by 48 h incubation, the medium was replaced with MTT
adhesion molecules including vascular adhesion molecule 1 (Sigma-Aldrich) dissolved at a final concentration of 1 mg/
(VCAM-1), intracellular adhesion molecule 1 (ICAM-1) and mL in serum-free, PR-free medium, for a further 4 h incu-
E-selectin on the surface of the vascular endothelium. The bation. Then, the MTT-formazan product was solubilized
increased expression of adhesion molecules by endothelial thoroughly in isopropanol and the optical density was
cells in human atherosclerotic lesions, activated by inflam- measured at a test wavelength of 550 nm and a reference
matory cytokines, may lead to further recruitment of wavelength of 690 nm.
leukocytes to atherosclerotic sites thereby resulting in the
formation of atherosclerotic lesion, oxidative modification of
LDL and endothelial dysfunction [12]. 2.3 Cell ELISA
In this study, we examined the hypothesis that b-sitos-
terol could inhibit the expression of adhesion molecules To examine whether b-sitosterol could modify the expres-
(VCAM-1 and ICAM-1) and the attachment of monocytes in sion of VCAM-1 and ICAM-1, cell ELISA was conducted.
human aortic endothelial cells (HAECs). The impact Briefly, to measure the cell-surface expression of adhesion
of treatment with b-sitosterol on tumor necrosis factor-a molecules, HAEC monolayer in flat-bottomed 96-well plates
(TNF-a)-related signaling in HAEC, such as activation (at 80% confluence) were pretreated with b-sitosterol (0.1, 1,
of nuclear factor-kB (NFkB) phosphorylation, was also 10, 50 and 100 mM) for 18 h, then stimulated for 6 h at 371C
examined. with 1 ng/mL TNF-a (Sigma-Aldrich), after which the cells
were fixed with 0.1% glutaraldehyde in PBS for 30 min at
41C. Plates were blocked at 371C for 1 h with 5% skimmed
2 Materials and methods milk powder in PBS, this followed by an incubation at 41C
overnight with a primary monoclonal goat antibody against
2.1 Culture of HAEC human ICAM-1 or VCAM-1, at final concentration 2 mg/mL
in 5% skimmed milk PBS. Next, the plates were washed
HAEC were provided as cryopreserved cells by Clonetics three times with 0.1% Tween-20 in PBS and incubated with
(Cambrex, USA) and were grown in culture flasks at 371C in a horseradish peroxidase-conjugated rabbit anti-mouse IgG
a humidified 95% air-5% CO2 atmosphere in endothelial secondary antibody at a dilution of 1:5000 at room
cell basal medium (Clonetics, Cambrex) supplemented with temperature (RT) for 1 h. Subsequently, the plates were
fetal bovine serum (FBS, 2%), human epidermal growth washed three times with 0.1% Tween-20 in PBS and finally
factor (10 ng/mL), hydrocortisone (1.0 mg/mL), gentamicin the expression of cell adhesion molecules (CAMs) was
(50 mg/mL), amphotericin B (50 ng/mL) and bovine brain quantified by the addition of the peroxides substrate
extract (3 mg/mL). The growth medium was changed every o-phenylendiamine hydrochloride. As a positive control, we
other day until confluence. Cells under passage 8 were used used vitamin E (a-tocopherol, 20 mM), an antioxidant known
for this study. to exert its effect through modulation of cytokines, adhesion
molecules, mobilization of NFkB transcription factor and
interaction of immune cells with endothelial cells [14, 15].
2.2 [3-(4,5-dimethylthiazol-2-yl)-2,5- The absorption of each well was measured at 450 nm using
diphenyltetrazolium bromide] assay a microplate ELISA reader.

Cultures of HAEC were grown in endothelial cell growth

medium (Clonetics, Cambrex) supplemented with 2% FBS 2.4 Fluorescent labeling of monocytes
(Gibco BRL, Invitrogen, USA), in T-75 cm2 flasks at 371C,
95% humidity and 5% CO2 atmosphere. Subcultures were U937 cells (human monocytic cell line which exhibits many
carried out every 3–4 days using a trypsin 0.025% and EDTA characteristics of monocytes) (ECACC, UK) were fluores-

& 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mnf-journal.com
Mol. Nutr. Food Res. 2010, 54, 551–558 553

cently labeled with 20 ,70 -bis-(2-cabroxyethyl)-5-(and-6)- of interest without the need to prepare cell lysates and to
carboxy-fluorescein acetoxymethyl ester (BCECF-AM; Mole- perform a Western blot. In our study, NFkB (p50/p65) is a
cular Probes, Invitrogen, USA) for the quantitative cell key transcription factor which is implicated in the regulation
adhesion assay. Non-fluorescent BCECF-AM is lipophilic of a variety of genes participating in inflammatory respon-
and when it is cleaved intercellularly it becomes a highly ses, including genes encoding VCAM-1 and ICAM-1. TNF-a
charged fluorescent BCECF, which is retained by viable induces the activation-phosphorylation and translocation of
cells. The BCECF-AM was obtained as a 1 g/L stock solution NFkB to the nucleus, promoting the VCAM-1 and ICAM-1
in anhydrous DMSO and was stored at -801C. After labeling expression. Thus, we first detected the time course of TNF-a
the U937 cells (106 cells/mL) with 10 mM BCECF-AM in induced phosphorylation on Ser-536 of NFkB p65, and
RPMI-1640 PR(-) (Gibco BRL, Invitrogen), FBS(-) and second we tested the impact on NFkB p65 phosphorylation
L-glutamine(-) medium for 1 h at 371C and 5% CO2, the cells b-sitosterol at the indicated concentrations. Briefly, the cells
were washed two times with PBS to remove the excess dye. were fixed with 4% fixing buffer for 20 min, RT. Next, plates
Finally, the cells were resuspended in RPMI-1640 PR(1), were blocked for 1 h, RT, with protein-based blocking buffer.
2% FBS medium at a density of 2  104 cells/well, according After blocking, incubation with the primary antibody for 1 h,
to the manufacturer’s instructions. RT, followed. The plates were then washed three times with
1  washing buffer and incubated with the secondary
antibody for 1 h, RT. Subsequently, the plates were washed
2.5 U937 cell adhesion assay three times with 1  washing buffer and finally colorimetric
detection of antibodies was carried out by measuring the
HAEC were cultured until confluence in 96-well plates and absorption of each well at 450 nm. Determination of relative
pretreated with b-sitosterol (0.1, 1, 10, 50 and 100 mM) for cell number was followed by reading the absorption of each
18 h. Then cells were stimulated for 6 h at 371C with TNF-a well at 595 nm. The absorption was measured by using a
(2 ng/mL). BCECF-labeled U937 cells (2  104 cells/well) microplate ELISA reader.
were incubated with HAEC for 1 h at 371C and 5% CO2.
After incubation, non-adherent cells were removed by
washing gently each well two times with PBS. The attached 2.8 Statistical analysis
cells were lysed with 50 mmol/L Tris buffer (pH 7.6)
containing 0.1% sodium dodecyl sulphate. a-Tocopherol Data are reported as mean7SD of three independent
was used as a positive control (20 mM). The fluorescent experiments (each experiment was conducted in triplicate or
intensity of each well was measured with a fluorescence quadruplicate). Data in figures are expressed as percentage
multiwell plate reader set at excitation and emission wave- of control, which was calculated as follows: (value for cells
lengths of 485 and 530 nm, respectively. treated with test compound/value for control cells)  100.
Statistical analysis was performed using Student’s t-test,
two-tailed distribution, assuming two-sample unequal
2.6 Blocking antibody studies variance.

mAbs against VCAM-1 or ICAM-1 were incubated with

confluent HAEC in 96-well plates at 5 mg/mL final concen- 3 Results
tration, during the final 30 min of the TNF-a stimulation,
before U937 were added [16, 17]. Adhesion assay was then 3.1 b-Sitosterol inhibits CAMs expression in HAEC
performed in sections 2.4, 2.5.
The effect of different concentrations of TNF-a (1 or 2 ng/
mL) on VCAM-1 and ICAM-1 expression was initially
2.7 Measurement of NFkB p65 phosphorylation determined after 6-, 12- or 24-h incubation. Incubation of
confluent HAEC with TNF-a (1 ng/mL) caused a maximal
To measure NFkB phosphorylation, confluent HAEC were surface expression of VCAM-1 and ICAM-1 after 6 h of
starved in serum free culture medium for 18 h and then incubation (data not shown). In subsequent experiments, we
pretreated without or with b-sitosterol (0.1, 1, 10, 50, 100 used TNF-a (1 ng/mL) for 6 h to induce stimulation of cells.
and 200 mM) for 18 h. Cells were then stimulated for 5 min TNF-a increased the basal expression (control) of CAMs
at 371C with TNF-a (2 ng/mL). To arrive at these conditions, VCAM-1 and ICAM-1 of confluent HAEC. a-Tocopherol
we performed a time course experiment by stimulating decreased significantly the TNF-a-induced endothelial
HAEC with TNF-a (2 ng/mL) for 5, 15 and 30 min. Finally, expression of both VCAM-1 and ICAM-1 (po0.001), as
NFkB p65 phosphorylation was measured with CASETM kit expected [18]. The effects of b-sitosterol on the expression of
(FE-005, SABiosciences, USA). The CASETM kit can be used VCAM-1 and ICAM-1 by HAEC are shown in Figs. 1A and
to monitor activation of a signal transduction pathway by B. b-Sitosterol (50, 100 mM) caused a significant dose
measuring phosphorylation of a key mediator of a pathway dependent decrease in VCAM-1 and ICAM-1 expression

& 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mnf-journal.com
554 S. Loizou et al. Mol. Nutr. Food Res. 2010, 54, 551–558

Figure 2. (A) b-sitosterol has no significant effect on VCAM-1

Figure 1. (A) b-Sitosterol inhibits TNF-a-induced VCAM-1 protein
protein expression in HAEC when tested alone. HAEC were
expression in HAEC. HAEC were incubated as described in
incubated as described in Section 2, in the absence of TNF-a or
Section 2, in the absence of TNF-a or compounds (control), with
compounds (control), with a-tocopherol (a-Toc) (20 mM), or with
a-tocopherol (a-Toc) (20 mM), or with different concentrations of
different concentrations of b-sitosterol (0.1–100 mM) for 18 h
b-sitosterol (0.1–100 mM) for 18 h, followed by stimulation with
(without stimulation with TNF-a). Adhesion molecules were
TNF-a (1 ng/mL) for up to 24 h. Adhesion molecules were
measured by cell ELISA. Data are expressed as percentage of
measured by cell ELISA. Data are expressed as percentage of
control and shown as means7SD of three independent experi-
control and shown as means7SD of three independent experi-
ments (each conducted in triplicates). A po0.05 value was
ments (each conducted in triplicates). A po0.05 value was
considered statistically significant when compared with control
considered statistically significant when compared to TNF-a-
(po0.01, po0.001). (B) b-Sitosterol has no significant effect
treated cells (po0.01, po0.001). (B) b-Sitosterol inhibits
on ICAM-1 protein expression in HAEC when tested alone. HAEC
TNF-a-induced ICAM-1 protein expression in HAEC. HAEC were
were incubated as described in Section 2, in the absence of
incubated as described in Section 2, in the absence of TNF-a or
TNF-a or compounds (control), with a-tocopherol (a-Toc)
compounds (control), with a-tocopherol (a-Toc) (20 mM), or with
(20 mM), or with different concentrations of b-sitosterol
different concentrations of b-sitosterol (0.1–100 mM) for 18 h,
(0.1-100 mM) for 18 h (without stimulation with TNF-a). Adhesion
followed by stimulation with TNF-a (1 ng/mL) for up to 24 h.
molecules were measured by cell ELISA. Data are expressed as
Adhesion molecules were measured by cell ELISA. Data are
percentage of control and shown as means7SD of three inde-
expressed as percentage of control and shown as means7SD of
pendent experiments (each conducted in triplicates). A po0.05
three independent experiments (each conducted in triplicates). A
po0.05 value was considered statistically significant when value was considered statistically significant when compared to
control (po0.01, po0.001).
compared with TNF-a-treated cells (po0.01, po0.001).

conditions used. The lowering effect of b-sitosterol on the

(po0.001), compared to TNF-a-treated HAEC. Furthermore, expression of adhesion molecules without affecting the
when we performed the same procedure, but without the viability rate of cells supports their anti-inflammatory activity.
stimulation of the cells with TNF-a, b-sitosterol caused a
small increase in VCAM-1 expression (1, 10 and 50 mM) and
had no effect on ICAM-1 expression (in all concentrations 3.3 b-Sitosterol inhibits binding of U937 cells to
tested), compared with control (Figs. 2A and B). TNF-a-stimulated HAEC

The effects of the test compounds on the binding of U937

3.2 Cell viability cells to TNF-a-stimulated HAEC were determined and results
are shown in Fig. 3. Control confluent HAEC showed
The assessment of cell viability revealed that neither the minimal binding to U937 cells and results are expressed as
morphology nor the reduction of MTT salt in HAEC cells was percentage of control. Adhesion was significantly increased
affected by any of the tested compounds (b-sitosterol, a-toco- when the HAEC were treated with TNF-a (2 ng/mL).
pherol or TNF-a) in any concentration range or experimental Pretreatment with a-tocopherol (20 mM) or b-sitosterol (1, 10,

& 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mnf-journal.com
Mol. Nutr. Food Res. 2010, 54, 551–558 555

50 and 100 mM) significantly (po0.01) reduced the adhesion

of U937 cells to TNF-a-stimulated HAEC.

3.4 Effect of antibodies against adhesion molecules

on U937 adhesion to HAEC

By using mAbs against human VCAM-1 or ICAM-1, we

evaluated their relative involvement in TNF-a induced
monocyte adhesion to HAEC. TNF-a-stimulated HAEC
(2 ng/mL), pretreated with b-sitosterol (100 mM) or not, were
incubated with anti-ICAM-1 or anti-VCAM-1 antibodies at
Figure 4. Blocking antibody studies. HAEC were incubated as
5 mg/mL for 30 min prior to the adhesion assay. As shown in described in Section 2, in the absence of TNF-a or compounds
Fig. 4, anti-VCAM-1 mAb alone inhibited significantly (control), with a-tocopherol (a-Toc) (20 mM), or with b-sitosterol
(po0.05) the number of adherent monocytes. Co-incubation (100 mM) for 18 h, followed by stimulation with TNF-a (2 ng/mL)
of anti-VCAM-1 antibody with 100 mM of b-sitosterol for up to 24 h. During the last 30 min of TNF-a stimulation and
enhanced the inhibition of U937 monocyte adhesion in just before conducting the U937 adhesion assay, mAb to human
VCAM-1 or ICAM-1 were added (5 mg/mL). Monocyte adhesion
TNF-a-stimulated HAEC (po0.01). Interestingly, anti-
was determined by the adhesion assay. Data are expressed as
ICAM-1 mAb treatment of HAEC or co-incubation of anti- percentage of control and shown as means7SD of two inde-
ICAM-1 with 100 mM of b-sitosterol markedly increased pendent experiments (each conducted in triplicates). A po0.05
U937 cell adhesion to HAEC (Fig. 4). value was considered statistically significant when compared to
TNF-a-treated cells (po0.01, po0.001).

3.5 b-Sitosterol attenuates phosphorylation of NFkB

p65 in TNF-a-stimulated HAEC

The time course experiment we performed to monitor the

NFkB phosphorylation, by stimulating HAEC with TNF-a
(2 ng/mL) for 5, 15 and 30 min, showed that stimulation of
the cells for 5 min was more significant (Fig. 5). Results
showed that pretreatment with b-sitosterol decrease signif-
icantly (po0.001) the phosphorylation of NFkB p65 in TNF-
a-stimulated HAEC compared with control HAEC (Fig. 6).
Figure 5. Monitoring of NFkB phosphorylation over a time
course. HAEC were grown in 96-well plates and starved before
treatment. Cells were then treated with 2 ng/mL of NFkB activator
TNF-a for 5, 15 and 30 min. Phosphorylation levels at Serine-536
of NFkB p65 were measured using the CASETM kit for NFkB S536
(Cat. No. FE-005).

4 Discussion

Accumulating evidence indicates that phytosterols exert

cardiovascular protective effects mainly via their cholesterol
lowering ability, modulation of endothelial function and
antioxidant capacity [19, 20]. More important, the cardio-
Figure 3. b-Sitosterol and a-tocopherol (a-Toc) inhibit TNF-a- vascular protective effect of b-sitosterol, a key component of
induced monocyte adhesion to HAEC. HAEC were incubated as cholesterol controlling functional foods, has been related to
described in Section 2, in the absence of TNF-a or compounds its antioxidant and hypocholesterolemic capacity [1].
(control), with a-tocopherol (a-Toc) (20 mM), or with different Data, however, on the role of b-sitosterol in the inflam-
concentrations of b-sitosterol (0.1-100 mM) for 18 h, followed by matory process of atherosclerosis are sparse. Since the
stimulation with TNF-a (2 ng/mL) for up to 24 h. Monocyte
binding and recruitment of circulating monocytes to
adhesion was determined by the adhesion assay. Data are
expressed as percentage of control and shown as means7SD of
vascular endothelial cells are early steps in the development
three independent experiments (each conducted in triplicates). A of inflammation and atherosclerosis, mediated through
po0.05 value was considered statistically significant when CAMs that are expressed on the surface of endothelial cells,
compared with TNF-a-treated cells (po0.01, po0.001). we evaluated the potential of b-sitosterol to influence the

& 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mnf-journal.com
556 S. Loizou et al. Mol. Nutr. Food Res. 2010, 54, 551–558

tory conditions, whereas at basal conditions it may induce

inflammatory processes. However, further studies are
warranted to elucidate the clinical impact of this observa-
tion, i.e. whether b-sitosterol supplementation in a healthy
population may be a risk factor for low-grade inflammation
and atherogenesis.
More importantly, the adhesion assay showed that in the
presence of b-sitosterol (0.1–100 mM), the binding of human
monocytic cell line, U937, to TNF-a-stimulated HAEC was
inhibited significantly. We further investigated the role of
VCAM-1 and ICAM-1 expression in mediating the inhibi-
Figure 6. b-Sitosterol inhibits TNFa-induced phosphorylation of
NFkB. HAEC were grown in 96-well plates and starved before
tory effects of b-sitosterol on adhesion of U937 cells to
treatment with test compound (0.1–200 mM). Cells were then HAEC. We examined the effects of blocking antibodies
treated with 2 ng/mL of NFkB activator TNF-a (2 ng/mL) for 5 min. (anti-VCAM-1 and anti-ICAM-1) on adhesion of U937 to
Phosphorylation levels at Serine-536 of NFkB p65 were HAEC stimulated with TNF-a and found that the use of
measured using the CASETM kit for NFkB S536. Data are anti-VCAM-1 inhibited significantly the U937 cell adhesion,
expressed as relative amount of phospho-NFkB and shown as whereas anti-ICAM-1 was without effect. Such data suggest
means7SD of three independent experiments (each conducted
that VCAM-1 expression plays an important role in U937
in triplicates). A po0.05 value was considered statistically
significant when compared to TNF-a-treated cells (po0.01, adhesion to HAEC. The small decrease by b-sitosterol in
po0.001). enhanced ICAM-1 expression induced by TNF-a (as shown
in cell ELISA) thus seems to be of no significance in U937
adhesion to HAEC. Interestingly, a combination of b-sitos-
expression of VCAM-1 and ICAM-1 by HAEC. We used the terol and anti-VCAM-1 antibody increased the magnitude of
Cell ELISA to measure VCAM-1 and ICAM-1, a well- inhibition of U937 cell adhesion to a greater degree than the
recognized in vitro assay, to evaluate the anti-inflammatory individual anti-VCAM-1 antibody. The degree of inhibition
effect of test compounds [21–23]. As a positive control, was similar to that observed by b-sitosterol alone. This may
we used vitamin E (a-tocopherol), an antioxidant known to be due to the expression of other available adhesion mole-
exert its effect through modulation of cytokines, cules on the endothelial surface being occupied by U937
adhesion molecules, mobilization of NFkB transcription cells, which molecules, in the presence of b-sitosterol,
factor and interaction of immune cells with endothelial cells become unavailable thus resulting in attenuation of U937
[14, 15]. For further evaluation of the anti-inflammatory binding. It is important to note that the combination of anti-
effect of b-sitosterol, we used the adhesion assay to measure ICAM-1 antibody with b-sitosterol attenuated the inhibitory
the monocyte adhesion to TNF-a-stimulated HAEC [24]. effect of b-sitosterol alone in U937 cell adhesion to HAEC.
To decide the proper dosages of b-sitosterol to be tested Previous reports have shown that ICAM-1 adhesion mole-
in our in vitro systems, we considered it important to take cules, which are expressed in a wide variety of cell types
into account the following: (i) b-sitosterol, the major including U937 [26], may use ICAM-1 antibody (bound to
phytosterol of higher plants, is found in the tissues and ICAM-1 molecules expressed on HAEC) as a bridge to
plasma of healthy individuals at concentrations 800–1000 facilitate further binding of several U937 cells [16]. In view
times lower than that of endogenous cholesterol [25] of the above, we speculate that similar interactions may have
and (ii) a recent report demonstrates that plasma concen- taken place in our study, thus masking the inhibitory effect
trations of b-sitosterol range from 2.8 to 16 mmol/L [1]. In of b-sitosterol alone in U937 cell adhesion to HAEC. Taken
view of the above, we decided to assess the biological together, our study demonstrates that b-sitosterol inhibits
effects of b-sitosterol at a wide concentration range from 0.1 TNF-a induced U937 cell adhesion to HAEC by lowering the
to 200 mM. Vitamin E inhibited the TNF-a induced expression of various adhesion molecules, of which the
expression of both ICAM-1 and VCAM-1, as expected [14, suppression of VCAM-1 adhesion molecule plays a key role
18]. The inhibition of TNF-a induced endothelial in U937 adhesion to HAEC.
activation and expression of ICAM-1 and VCAM-1 adhesion Summarizing, since monocyte recruitment into the
molecules by b-sitosterol, at a concentration range vascular wall, after their adhesion to endothelial cells, is a
0.1–100 mM, implicates the anti-inflammatory effect of b- crucial step in the pathogenesis of atherosclerosis, our data
sitosterol on endothelial cells. Meanwhile, b-sitosterol, when strongly indicate that b-sitosterol has a noticeable anti-
tested alone (in the absence of HAEC stimulation with atherogenic potential. In agreement with the present find-
TNF-a), had no significant action (compared with control). ings, animal studies have concluded that PS (mixture of:
In our study, the small increase in VCAM-1 expression b-sitosterol, campesterol and stigmasterol) exerts protective
by b-sitosterol treatment without TNF-a stimulation effects on atherosclerotic lesion development, plaque forma-
suggests that the lowering effect of b-sitosterol on adhesion tion, foam cell formation and vascular endothelium damage
molecule expression is more pronounced under inflamma- [27]. More recently, Bustos et al. showed that b-sitosterol

& 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mnf-journal.com
Mol. Nutr. Food Res. 2010, 54, 551–558 557

inhibited expression of ICAM-1 and decreased migration and In conclusion, our study extends existing data regarding
adhesion of THP-1 cells to oxLDL-stimulated HUVEC [7]. the cardioprotective effect of b-sitosterol and provides new
Furthermore, the excessive production of reactive oxygen insights into understanding the molecular mechanism
species contributes to the pathogenesis of cardiovascular underlying the beneficial effect of b-sitosterol on endothelial
disease. To this effect, the cardioprotection of b-sitosterol, function and cardioprotection.
mechanistically, has been shown to be mediated via its ability
to strengthen the intracellular antioxidant defense. In parti- The authors have declared no conflict of interest.
cular, b-sitosterol has been shown to protect against oxidative
stress via modulation of antioxidant enzymes [28]. More
importantly, a reduction by b-sitosterol in arachidonic acid
release and prostanglandin E2 production, induced by 5 References
oxidized low-density lipoproteins, has been observed [29].
[1] Chan, Y. M., Varady, K. A., Lin, Y., Trautwein, E. et al.,
Such data implicate possible mechanism(s) involved in the
Plasma concentrations of plant sterols: physiology and
regulatory effects of b-sitosterol in adhesion molecule
relationship with coronary heart disease. Nutr. Rev. 2006,
expression, cell-cell interaction and inflammation [30]. 64, 385–402.
However, De Jong et al. [31] demonstrated that b-sitosterol did
[2] St-Onge, M. P., Jones, P. J., Phytosterols and human lipid
not modify soluble adhesion molecules, neither did it affect
metabolism: efficacy, safety, and novel foods. Lipids 2003,
markers of antioxidant status or oxidative stress, despite a 38, 367–375.
significant reduction in LDL cholesterol in patients treated
[3] Katan, M. B., Grundy, S. M., Jones, P., Law, M. et al., Effi-
with statins. Such observations suggest that under in vivo
cacy and safety of plant stanols and sterols in the
conditions, the combination of therapeutic regimes may management of blood cholesterol levels. Mayo Clin. Proc.
result in synergistic or antagonistic effects and thus variable 2003, 78, 965–978.
biological activity. The clinical impact of such effects should
[4] Thompson, G. R., Additive effects of plant sterol and stanol
be taken into consideration. esters to statin therapy. Am. J. Cardiol. 2005, 96, 37D–39D.
To further elucidate the mechanism of action of b-sitos-
[5] Grundy, S. M., Stanol esters as a component of maximal
terol, we investigated their effects on the phosphorylation of
dietary therapy in the National Cholesterol Education
NFkB p65, which is a key transcription factor regulating a Program Adult Treatment Panel III report. Am. J. Cardiol.
variety of genes participating in inflammatory responses, 2005, 96, 47D–50D.
including genes encoding VCAM-1 and ICAM-1 [32]. NFkB
[6] Nashed, B., Yeganeh, B., HayGlass, K. T., Moghadasian,
is a cytoplasmatic component which is an inactive complex M. H., Antiatherogenic effects of dietary plant sterols
with its inhibitor IkB. Its activation is the critical process for are associated with inhibition of proinflammatory
the downstream activation and gene expression in endo- cytokine production in Apo E-KO mice. J. Nutr. 2005, 135,
thelial cells. TNF-a can induce the phosphorylation of IkB, 2438–2444.
which results in dissociation of IkB and finally activation- [7] Bustos, P., Duffau, C., Pacheco, C., Ulloa, N., Beta-sitosterol
phosphorylation and translocation of NFkB to the nucleus, modulation of monocyte-endothelial cell interaction: A
promoting the expression of downstream genes, such as comparison to female hormones. Maturitas 2008, 60,
adhesion molecules VCAM-1 and ICAM-1. 202–208.
In our study, we found that TNF-a treatment induced [8] De la Puerta, R., Martỳnez-Domingues, E., Ruiz-Gutierrez, V.,
higher levels of NFkB phosphorylation in HAEC, an indi- Effect of minor components of virgin olive oil on topical
cation that NFkB was activated. Treatment with b-sitosterol anti-inflammatory assays. Z. Naturforsch. [C] 2000, 55,
(0.1–200 mM) significantly inhibited the phosphorylation 814–819.
of NFkB, suggesting that its anti-inflammatory activity in [9] Navarro, A., De las Heras, B., Villar, A., Antiinflammatory
vitro is mediated, at least in part, via the inactivation of and immunomodulating properties of a sterol fraction from
NFkB. In the same line with our findings, Moreno showed Sideritis foetens. Chem. Biol. Pharm. Bull. 2001, 24,
that b-sitosterol decreased the activation of NFkB tran- 470–473.
scription factor in PMA-stimulated macrophage cells [10]. [10] Moreno, J. J., Effect of olive oil minor components on
Although it is widely recognized that induction of endo- oxidative stress and arachidonic acid mobilization and
thelial adhesion molecules by inflammatory cytokines metabolism by macrophages RAW 264.7. Free Radic. Biol.
depends greatly on NFkB activation and its DNA-binding Med. 2003, 35, 1073–1081.
ability, however, the promoters of adhesion molecules, [11] Han, S. H., Quon, M. J., Koh, K. K., Reciprocal relationships
including ICAM-1 and VCAM-1 promoters, contain binding between abnormal metabolic parameters and endothelial
sites for other transcription factors such as AP-1 or SP-1 dysfunction. Curr. Opin. Lipidol. 2007, 18, 58–65.
[32–36]. In view of the above, the anti-inflammatory effect of [12] Ross, R., Atherosclerosis – an inflammatory disease. N. Eng.
b-sitosterol on HAEC may be mediated by activation of J. Med. 1999, 340, 115–126.
multiple transcription factors and certainly warrants further [13] Denizot, F., Lang, R., Rapid colorimetric assay for cell
investigation. growth and survival. Modifications to the tetrazolium

& 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mnf-journal.com
558 S. Loizou et al. Mol. Nutr. Food Res. 2010, 54, 551–558

dye procedure giving improved sensitivity and reliability. [25] Bouic, P. J., The role of phytosterols and phytosterolins in
J. Immunol. Methods 1986, 89, 271–277. immune modulation: a review of the past 10 years. Curr.
[14] Wu, D., Koga, T., Martin, K. R., Meydani, M., Effect of vita- Opin. Clin. Nutr. Metab. Care 2001, 4, 471–475.
min E on human aortic endothelial cell production of [26] Dustin, M. L., Rothlein, R., Bhan, A. K., Dinarello, C. A. et al.,
chemokines and adhesion to monocytes. Atherosclerosis Induction by IL 1 and interferon-gamma: tissue distribution,
1999, 147, 297–307. biochemistry, and function of a natural adherence molecule
[15] Weber, C., Erl, W., Pietsch, A., Strobel, M. et al., Antioxidants (ICAM-1). J. Immunol. 1986, 137, 245–254.
inhibit monocyte adhesion by suppressing nuclear factor- [27] Moghadasian, M. H., McManus, B. M., Pritchard, P. H.,
kappa B mobilization and induction of vascular cell adhesion Frohlich, J. J., ‘‘Tall oil’’-derived phytosterols reduce
molecule-1 in endothelial cells stimulated to generate radi- atherosclerosis in ApoE-deficient mice. Arterioscler.
cals. Arterioscler. Thromb. 1994, 14, 1665–1673. Thromb. Vasc. Biol. 1997, 17, 119–126.
[16] Koga, T., Claycombe, K., Meydani, M., Homocysteine [28] Vivancos, M., Moreno, J. J., beta-Sitosterol modulates
increases monocyte and T-cell adhesion to human aortic antioxidant enzyme response in RAW 264.7 macrophages.
endothelial cells. Atherosclerosis 2002, 161, 365–374. Free Radic. Biol. Med. 2005, 39, 91–97.
[17] Silverman, M. D., Tumuluri, R. J., Davis, M., Lopez, G. et al., [29] Vivancos, M., Moreno, J. J., Effect of resveratrol, tyrosol
Homocysteine upregulates vascular cell adhesion mole- and beta-sitosterol on oxidised low-density lipoprotein-
cule-1 expression in cultured human aortic endothelial cells stimulated oxidative stress, arachidonic acid release and
and enhances monocyte adhesion. Arterioscler. Thromb. prostaglandin E2 synthesis by RAW 264.7 macrophages. Br.
Vasc. Biol. 2002, 22, 587–592. J. Nutr. 2008, 99, 1199–1207.
[18] Zapolska-Downar, D., Zapolski-Downar, A., Markiewski, M., [30] Moreno, J. J., Antiflammin-2 prevents HL-60 adhesion to
Ciechanowicz, A. et al., Selective inhibition by alpha-toco- endothelial cells and prostanoid production induced by
pherol of vascular cell adhesion molecule-1 expression in lipopolysaccharides. J. Pharmacol. Exp. Ther. 2001, 296,
human vascular endothelial cells. Biochem. Biophys. Res. 884–889.
Commun. 2000, 274, 609–615. [31] De Jong, A., Plat, J., Bast, A., Godschalk, R. W. et al., Effects
[19] Marinangeli, C. P., Varady, K. A., Jones, P. J., Plant sterols of plant sterol and stanol ester consumption on lipid
combined with exercise for the treatment of hypercholes- metabolism, antioxidant status and markers of oxidative
terolemia: overview of independent and synergistic stress, endothelial function and low-grade inflammation in
mechanisms of action. J. Nutr. Biochem. 2006, 17, 217–224. patients on current statin treatment. Eur. J. Clin. Nutr. 2008,
[20] Patel, M. D., Thompson, P. D., Phytosterols and vascular 62, 263–273.
disease. Atherosclerosis 2006, 186, 12–19. [32] Collins, T., Read, M. A., Neish, A. S., Whitley, M. Z. et al.,
[21] Kaneko, M., Hayashi, J., Saito, I., Miyasaka, N., Probucol Transcriptional regulation of endothelial cell adhesion
downregulates E-selectin expression on cultured human molecules: NF-kappa B and cytokine-inducible enhancers.
vascular endothelial cells. Arterioscler. Thromb. Vasc. Biol. FASEB J. 1995, 9, 899–909.
1996, 16, 1047–1051. [33] Neish, A. S., Williams, A. J., Palmer, H. J., Whitley, M. Z. et al.,
[22] Wolle, J., Hill, R. R., Ferguson, E., Devall, L. J. et al., Functional analysis of the human vascular cell adhesion
Selective inhibition of tumor necrosis factor-induced molecule 1 promoter. J. Exp. Med. 1992, 176, 1583–1593.
vascular cell adhesion molecule-1 gene expression by a [34] Neish, A. S., Khachigian, L. M., Park, A., Baichwal, V.R. et al.,
novel flavonoid. Lack of effect on transcription factor Sp1 is a component of the cytokine-inducible enhancer in
NF-kappa B. Arterioscler. Thromb. Vasc. Biol. 1996, 16, the promoter of vascular cell adhesion molecule-1.
1501–1508. J. Biol. Chem. 1995, 270, 28903–28909.
[23] Zhang, W. J., Frei, B., Albumin selectively inhibits TNF [35] Simoncini, T., Maffei, S., Basta, G., Barsacchi, G. et al.,
alpha-induced expression of vascular cell adhesion mole- Estrogens and glucocorticoids inhibit endothelial vascular
cule-1 in human aortic endothelial cells. Cardiovasc. Res. cell adhesion molecule-1 expression by different transcrip-
2002, 55, 820–829. tional mechanisms. Circ. Res. 2000, 87, 19–25.
[24] Kim, J. D., Liu, L., Guo, W., Meydani, M., Chemical structure [36] Zhang, W. J., Frei, B., Intracellular metal ion chelators
of flavonols in relation to modulation of angiogenesis and inhibit TNFalpha-induced SP-1 activation and adhesion
immune-endothelial cell adhesion. J. Nutr. Biochem. 2006, molecule expression in human aortic endothelial cells. Free
17, 165–176. Radic. Biol. Med. 2003, 34, 674–682.

& 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mnf-journal.com