Stevioside inhibits atherosclerosis by improving insulin signaling and antioxidant defense in obese insulin-resistant mice

Abstract

Stevioside is a non-caloric natural sweetener that does not induce a glycemic response, making it attractive as sweetener to diabetics and others on carbohydrate-controlled diets. Obesity is frequently associated with insulin resistance and increased inflammation and oxidative stress. Therefore, we investigated its effects on insulin resistance, inflammation and oxidative stress related to atherosclerosis in obese insulin-resistant mice.
Twelve-week-old mice were treated with stevioside (10 mg kg(-1), n=14) or placebo (n=20) for 12 weeks.
Stevioside had no effect on weight and triglycerides, but lowered glucose and insulin. Stevioside treatment improved adipose tissue maturation, and increased glucose transport, insulin signaling and antioxidant defense in white visceral adipose tissues. Together, these increases were associated with a twofold increase of adiponectin. In addition, stevioside reduced plaque volume in the aortic arch by decreasing the macrophage, lipid and oxidized low-density lipoprotein (ox-LDL) content of the plaque. The higher smooth muscle cell-to-macrophage ratio was indicative for a more stable plaque phenotype. The decrease in ox-LDL in the plaque was likely due to an increase in the antioxidant defense in the vascular wall, as evidenced by increased Sod1, Sod2 and Sod3. Circulating adiponectin was associated with improved insulin signaling and antioxidant defense in both the adipose tissue and the aorta of stevioside-treated mice.
Stevioside treatment was associated with improved insulin signaling and antioxidant defense in both the adipose tissue and the vascular wall, leading to inhibition of atherosclerotic plaque development and inducing plaque stabilization.

Full text

ORIGINAL ARTICLE
Stevioside inhibits atherosclerosis by improving
insulin signaling and antioxidant defense in obese
insulin-resistant mice
B Geeraert
1
, F Crombe
´
1
, M Hulsmans
1
, N Benhabile
`s
1
, JM Geuns
2
and P Holvoet
1
1
Atherosclerosis and Metabolism Unit, Department of Cardiovascular Diseases and Leuven Food Science and Nutrition
Research Centre, Katholieke Universiteit Leuven, Leuven, Belgium and
2
Laboratory of Functional Biology, Katholieke
Universiteit Leuven, Leuven, Belgium
Objective: Stevioside is a non-caloric natural sweetener that does not induce a glycemic response, making it attractive as
sweetener to diabetics and others on carbohydrate-controlled diets. Obesity is frequently associated with insulin resistance and
increased inflammation and oxidative stress. Therefore, we investigated its effects on insulin resistance, inflammation and
oxidative stress related to atherosclerosis in obese insulin-resistant mice.
Research design: Twelve-week-old mice were treated with stevioside (10 mg kg
1
,n¼14) or placebo (n¼20) for 12 weeks.
Results: Stevioside had no effect on weight and triglycerides, but lowered glucose and insulin. Stevioside treatment improved
adipose tissue maturation, and increased glucose transport, insulin signaling and antioxidant defense in white visceral adipose
tissues. Together, these increases were associated with a twofold increase of adiponectin. In addition, stevioside reduced plaque
volume in the aortic arch by decreasing the macrophage, lipid and oxidized low-density lipoprotein (ox-LDL) content of the
plaque. The higher smooth muscle cell-to-macrophage ratio was indicative for a more stable plaque phenotype. The decrease in
ox-LDL in the plaque was likely due to an increase in the antioxidant defense in the vascular wall, as evidenced by increased
Sod1,Sod2 and Sod3. Circulating adiponectin was associated with improved insulin signaling and antioxidant defense in both
the adipose tissue and the aorta of stevioside-treated mice.
Conclusion: Stevioside treatment was associated with improved insulin signaling and antioxidant defense in both the adipose
tissue and the vascular wall, leading to inhibition of atherosclerotic plaque development and inducing plaque stabilization.
International Journal of Obesity (2010) 34, 569–577; doi:10.1038/ijo.2009.261; published online 15 December 2009
Keywords: atherosclerosis; oxidative stress; insulin signaling; natural sweetener
Introduction
The growing incidence of obesity and obesity-related
cardiovascular risk factors have led to the quest for natural
sweeteners that can substitute for sucrose, and do not
provide calories. Much attention has been placed on glyco-
sides that are extracted from Stevia rebaudiana Bertoni. The
most abundant glycoside is stevioside, which is one of the
major sweeteners in use in Japan and Korea.
1
The increasing prevalence of obesity, a low-grade inflam-
matory and oxidative stress state, is closely associated with
the rising incidence of type diabetes and cardiovascular
diseases.
2
Indeed, recent studies showed a positive relation-
ship between obesity, inflammatory C-reactive protein,
3
and
oxidized low-density lipoprotein (ox-LDL).
4,5
Moreover, we
showed that ox-LDL was associated with the incidence of the
metabolic syndrome
6
and with several metabolic syndrome
components, especially visceral obesity.
7,8
Stevioside was found to exert antihyperglycemic and
insulinotropic effects in a non-obese animal model of type
2 diabetes
9
by acting directly on pancreatic cells.
10
However,
its effects on adipose tissue and obesity-associated insulin
resistance, inflammation, oxidative stress and atherosclerosis
have not been determined. Therefore, our aim was to
measure these effects and identify underlying molecular
pathways. We selected mice with combined leptin and
Received 26 August 2009; revised 27 October 2009; accepted 1 November
2009; published online 15 December 2009
Correspondence: Dr P Holvoet, Atherosclerosis and Metabolism Unit,
Department of Cardiovascular Diseases, Katholieke Universteit Leuven,
Herestraat 49, O&N1, PB 705, Leuven 3000, Belgium.
E-mail: paul.holvoet@med.kuleuven.be
International Journal of Obesity (2010) 34, 569–577
&
2010 Macmillan Publishers Limited All rights reserved 0307-0565/10
$
32.00
www.nature.com/ijo

LDL-receptor deficiency (double knockout (DKO) mice).
These exhibit most of the metabolic syndrome components,
which are associated with increased inflammation and
oxidative stress, accelerated atherosclerosis and impaired
cardiovascular function.
11–13
Weight loss
11
and treatment
with statin
12
were associated with improved insulin sensi-
tivity and decreased macrophage and ox-LDL accumulation
in their aortic arch. These improvements were associated
with increased expression of the antioxidant superoxide
dismutases (Sods) in the aorta.These observations prompted
us to investigate the effect of stevioside on the expression of
factors involved in the regulation of adipose tissue matura-
tion, insulin signaling, inflammation and oxidative stress in
the adipose tissue and aorta in relation to atherosclerosis in
these mice. We found a relation between adipose tissue
maturation, adiponectin, insulin signaling and the anti-
oxidant defense in the adipose tissue and the aorta of
stevioside-treated mice.
Research design and methods
Experimental protocol of animal studies
Experimental procedures in animals were performed in
accordance with protocols approved by the Institutional
Animal Care and Research Advisory Committee. DKO mice,
on the C57BL6 background, were obtained as previously
described.
11,13
All offspring were genotyped by PCR techni-
ques.
14
Mice were treated with stevioside (n¼14) or placebo
(n¼20) for 12 weeks starting at the age of 12 weeks.
Stevioside (molecular weight: 804) was dissolved in physio-
logical saline solution (0.9% NaCl) (1 mg ml
1
) and adminis-
tered orally at a dose of 10 mg kg
1
day
1
. Stevioside was
purified as described previously with a 99.9% purity.
15,16
Placebo was the solvent without active compound. All mice
were housed at 22 1C on a fixed 12/12-hour light–dark cycle
and were fed standard chow containing 4% fat. Food and
water were available ad libitum. Food intake was E5.7 g day
1
and was not affected by the treatment.
Biochemical analyses
Blood from awaken mice was collected by tail bleeding into
EDTA tubes after an overnight fast. During killing, total
blood was collected by puncturing the vena cava. Plasma
was obtained by centrifugation. Total cholesterol and
triglycerides were measured with standard enzymatic assays
(Boehringer, Mannheim, Germany), glucose with a gluco-
meter (Menarini Diagnostics, Zaventem, Belgium) and
plasma insulin with a mouse ELISA (Mercodia, Oxon, UK).
To determine glucose tolerance, glucose was measured in
samples obtained by tail bleeding before and 15, 30, 60,
120 and 240 min after intraperitoneal glucose administra-
tion (20% glucose solution; 2 g kg
1
).
11,12
Adiponectin,
interleukin-6 and tumor necrosis factor-awere measured
with specific mouse ELISAs (R&D Systems, Uppsala, Sweden).
Because the assay for ox-LDL in blood is based on a mouse
monoclonal antibody, ox-LDL cannot be measured directly
in mouse blood. Therefore, we measured the titer of auto-
antibodies against malondialdehyde (MDA)-modified LDL
as a proxy for ox-LDL in mice, as described before.
13,17,18
Real-time reverse transcription-PCR analysis
The levels of RNA expression in extracts of white visceral (IP)
adipose tissue and aorta were measured by quantitative real-
time reverse transcription-PCR. Total RNA was extracted with
Trizol reagent (Invitrogen, Merelbeke, Belgium) and purified
on RNeasy Mini kit columns (Qiagen, Venlo, The Netherlands).
First-strand cDNA was generated from total RNA with
the SuperScript VILO cDNA Synthesis Kit (Invitrogen).
Quantitative real-time reverse transcription-PCR was per-
formed using Power SYBR Green Master mix according to the
supplier protocols (Applied Biosystems, Lennik, Belgium).
Oligonucleotides (Invitrogen) used as forward and reverse
primers were designed using the ‘Primer Express’ software
(Applied Biosystems) and are summarized in Table 1. Primer
sequences were validated for specificity by Basic Local
Alignment Search Tool (BLAST).
19
PCR fragments were
validated for GC/AT ratio, length and amplification specifi-
city with dissociation curve analysis and agarose gel electro-
phoresis.
20
The level of RNA expression was calculated
using the threshold cycle (C
t
) value, normalized with the
housekeeping gene, b-actin, and related to an external
calibrator consisting of extracts from intra-abdominal adipose
tissue or aorta of C57BL6 control mice. Subsequently, DDC
t
(DC
t,sample
–DC
t,calibrator
) was determined, and the relative
expression levels were calculated from 2
–DDCt
. RNA expression
levels are thus indicated as arbitrary units ±s.d.
11–13
Atherosclerosis
The extent of atherosclerosis was determined by analysis of
ten 7-mm cross-sections of aortic root of 24-week placebo-
and stevioside-treated DKO mice. Lipids were stained with
oil red O, ox-LDL with mAb4E6,
21
smooth muscle cells
(SMC) with an a-actin-specific antibody (Dako, Heverlee,
Belgium), macrophages with an antibody against mouse
Mac-3 antigen (Pharmingen, Erembodegem, Belgium) and
paraoxonase 1 (Pon1) with polyclonal antibodies from Santa
Cruz Biotechnology (Tebu-bio, Boechout, Belgium). A color
intensity threshold mask for immunoassaying was defined to
detect the red color by sampling, and the same threshold was
applied to all specimens. Blinded analysis was performed
with the Quantimet 600 image analyzer (Leica, Groot
Bijgaarden, Belgium). The positively immunostained area
was expressed as a percentage of the total plaque area.
11,12
Statistical analysis
Groups were compared by means of the unpaired t-test
with Welch’s Correction (Graph Pad Prism version 5; La Jolla,
Stevioside on atherosclerosis in obesity
B Geeraert et al
570
International Journal of Obesity

CA, USA). Correlations were calculated using the non-
parametric Spearman’s correlation coefficient (Rs). The area
under the curve of the glucose tolerance test was calculated
using Graph Pad Prism version 5. Po0.05 was considered to
be statistically significant.
Results
Weight and blood analysis
Stevioside treatment had no effect on weight. It lowered
blood glucose, insulin and cholesterol, but had no effect on
triglycerides or glucose tolerance. Treatment with stevioside
nearly doubled plasma adiponectin concentrations, but did
not change plasma interleukin-6 and tumor necrosis factor-a
concentrations. The titer of autoantibodies against MDA-
modified LDL was decreased in stevioside-treated mice
(Table 2) and correlated inversely with the plasma adipo-
nectin concentration (Rs ¼0.66; Po0.001).
Insulin signaling and oxidative stress in visceral adipose tissue
The RNA expressions of Insr,Irs1,Irs2,Glut4,Fabp4 and
Lxrain the visceral adipose tissue of placebo-treated DKO
mice were lower than those in adipose tissue of lean
control C57BL6 mice. Stevioside treatment increased their
expressions. However, all expressions were still lower in
stevioside-treated mice than in lean controls (Figure 1).
Table 2 Effect of stevioside on blood values
Blood variables Placebo (N¼20) Stevioside (N¼14) P-value
Weight (g) 62.7±5.1 62.9±3.6 NS
Total cholesterol (mmol l
1
) 13.51±3.40 10.71±2.61 **
Triglycerides (mmol l
1
) 3.24±1.51 2.61±1.22 NS
Glucose (mmol l
1
) 8.19±1.76 6.73±1.97 *
Insulin (mU l
1
) 1831±728 1201±307 *
AUC of IPGTT
a
86873±21317 86135±18747 NS
Adiponectin (mg ml
1
) 3296±1837 6517±1472 ***
IL-6 (pg ml
1
) 17.26±9.56 13.20±4.64 NS
TNF-a(pg ml
1
) 24.50±10.93 27.86±4.74 NS
Titers of autoantibodies against MDA-modified LDL 9.20±1.84 3.45±1.36 ***
Abbreviations: IL, interleukin; LDL, low-density lipoprotein; MDA, malondialdehyde; NS, not significant; TNF, tumor necrosis factor.
a
AUC of IPGTT ¼area under the
curve of the intraperitoneal glucose tolerance test. Data are means±s.d. *Po0.05, **Po0.01 and ***Po0.001 compared with placebo-treated DKO mice.
Table 1 Primers used for quantitative real-time reverse transcription-PCR analysis
Gene name Gene symbol Forward Reverse
ATP-binding cassette, sub-family A, member 1 Abca1 TGGGCAGTCCAATCTAATCCAT GGAGGGCATGTGGGAAGAA
Catalase Cat CCTCTCCAACAGGCAAGTTTTT GGCAGTCTATTGCAAGTTCCATT
Chemokine (C-C motif) ligand 2 Ccl2 or Mcp1 GCAGTTAACGCCCCACTCA CAGCCTACTCATTGGGATCATCTT
Chemokine (C-C motif) receptor 2 Ccr2 GGGATCATGACCCAAAGTAAGAA GGTCTCGGTTGGGTTGTAAAGT
CD36 antigen Cd36 CTCGGACATTGAGATTCTTTTCCT GTCGATTTCAGATCCGAACACA
CD44 antigen Cd44 GACCCCGGACCAGTGTATGA CCAATCGTGCTGTCTTTTCAAGT
CD68 antigen Cd68 GCACAGCCAGCCCTACGA GAGCTGGTGTGAACTGTGACATTT
Fatty acid-binding protein 4 Fabp4 AACTGGGCGTGGAATTCG CTAGGGTTATGATGCTCTTCACCTT
Solute carrier family 2
(facilitated glucose transporter), member 4
Glut4 or Slc2a4 CACTGCTTCTGGCTCTCACAGTAC GTTCCGGATGATGTAGAGGTATCTG
Intercellular adhesion molecule 1 Icam1 GGGACCACGGAGCCAATT GCTTTGGGATGGTAGCTGGAA
Insulin receptor Insr CATGTGCAGGAATGGCTTGTT TTCTGCGTTTTCTGCAGTGCTA
Insulin receptor substrate 1 Irs1 AGCCCAGTGAGTCTGTCATCTAGTAG CGCCTCGGGAAGAGACAGT
Insulin receptor substrate 2 Irs2 CGAAGTACTCGTCCTTGGTGTAGA CGAGAAGAAGTGGAGGAGCAA
Liver X-receptor-aLxraGGAGTGTCGACTTCGCAAATG TCAAGCGGATCTGTTCTTCTGA
Nuclear factor -B subunit 1 Nfkb1 GGATGACAGAGGCGTGTATTAGG GTAGATAGGCAAGGTCAGAATGCA
Nuclear factor of -light polypeptide gene enhancer
in B-cells inhibitor-a
NfkbiaACCTGCACACCCCAGCAT CGTGTGGCCATTGTAGTTGGT
Peroxisome proliferator-activated receptor-aPparaTCAGGGTACCACTACGGAGTTCA CCGAATAGTTCGCCGAAAGA
Peroxisome proliferator-activated receptor-gPpargGCAGCTACTGCATGTGATCAAGA GTCAGCGGGTGGGACTTTC
Superoxide dismutase 1 Sod1 GGGATTGCGCAGTAAACATTC AATGGTTTGAGGGTAGCAGATGA
Superoxide dismutase 2 Sod2 TCGGTGGCGTTGAGATTGT ACACATTAACGCGCAGATCATG
Superoxide dismutase 3 Sod3 CATGCAATCTGCAGGGTACAAC GCTGCCGGAAGAGAACCAA
b-Actin b-Actin ACGGCCAGGTCATCACTATTG CACAGGATTCCATACCCAAGAAG
Stevioside on atherosclerosis in obesity
B Geeraert et al
571
International Journal of Obesity

Stevioside treatment also increased plasma adiponectin that
correlated Insr, Irs1 and Irs2 (Figure 1).
Previously, obesity in DKO mice was shown to be
associated with increased oxidative stress due to a lack
of antioxidant enzymes.
13
Therefore, we measured Sod3,
because it was shown to have a protective role against
hyperglycemia-induced reactive oxygen species (ROS).
22
Sod3 was decreased in placebo-treated DKO mice compared
with lean control mice; stevioside normalized its expression
from 0.64±0.28 to 1.06±0.47 (Po0.05). In addition, stevio-
side increased the expression of Cat in DKO mice from
0.30±0.08 to 0.51±0.12 (Po0.001), important for further
conversion of H
2
O
2
, generated by SOD, to water. The
expressions of antioxidant enzymes correlated negatively
with the titer of autoantibodies against MDA-modified LDL
(Rs ¼0.41, Po0.05 for Sod3 and Rs ¼0.58, Po0.001 for
Cat). Furthermore, both Sod3 and Cat correlated strongly
with plasma adiponectin (Rs ¼0.778 and 0.741, respectively;
both Po0.001) and Irs2 (Rs ¼0.407, Po0.05 and Rs ¼0.530,
Po0.01, respectively).
Atherosclerosis
Figure 2 shows representative sections of the aortic arch of
placebo- and stevioside-treated DKO mice with lipids stained
with oil red O, macrophages with an antibody against Mac-3
and ox-LDL with 4E6. Stevioside inhibited atherosclerosis by
reducing macrophage, ox-LDL and lipids. Furthermore,
stevioside treatment increased the SMC area of the plaque.
This increase together with the reduction of macrophages
resulted in an increase of the SMC-to-macrophage ratio.
Plaque macrophages correlated with ox-LDL (Rs ¼0.33,
Po0.01) that correlated with lipids (Rs¼0.48, Po0.01).
Plaque ox-LDL correlated with the titer of autoantibodies
against MDA-modified LDL (Rs ¼0.45, Po0.001). There was
no difference in PON1 staining in control and stevioside-
treated mice (9.0±4.4 vs 6.5±4.1%).
Insulin signaling, oxidative stress and inflammation in the aorta
To get a better insight in the pathways that are involved in
decreasing the oxidative stress in the aorta of stevioside-
Figure 1 Effect of stevioside on RNA expressions in the visceral adipose tissue. The RNA expressions of Insr,Irs1,Irs2, Lxra,Fabp4 and Glut4 were measured by
quantitative real-time reverse transcription-PCR. The expression of all these genes was lower in double knockout (DKO) than in lean C57BL6 control mice (n¼12);
DKO mice have the C57BL6 genetic background. Stevioside increased all their expressions. Adiponectin correlated with Insr, Irs1 and Irs2 in the adipose tissue. Data
are means±s.e.m. **Po0.01 and ***Po0.001 compared with lean control mice (C57BL6); and
$
Po0.05,
$$
Po0.01 and
$$$
Po0.001 compared with placebo DKO
mice. Rs: Spearman’s correlation coefficients.
Stevioside on atherosclerosis in obesity
B Geeraert et al
572
International Journal of Obesity

treated DKO mice, we measured the RNA expression of Sods.
Figure 3 shows that all Sods were lower in the aorta of DKO
mice than in lean control mice and that stevioside increased
their expressions. All Sods were inversely related with the
ox-LDL content of the plaque (Figure 3). In addition, the titer
of autoantibodies against MDA-LDL, used as a proxy for
ox-LDL in the circulation, correlated negatively with Sod1
(Rs ¼0.49, Po0.05), Sod2 (Rs ¼0.51, Po0.01) and Sod3
(Rs ¼0.58, Po0.001). Previously, we found that overexpres-
sion of Sod depends on induction of Ppars, especially Pparg.
23
Ppargwas lower in the aorta of DKO mice, and stevioside
restored its expression (Figure 3). Ppargexpression correlated
with Sod1 (Rs ¼0.49, Po0.05), Sod2 (Rs ¼0.57, Po0.01) and
Sod3 (Rs ¼0.67, Po0.001). Adiponectin correlated with
Pparg,Sod1 and Sod3 (Figure 3). Stevioside treatment had
no effect on Ppara (data not shown).
Figure 2 Effect of stevioside on atherosclerosis in the aortic arch. Representative sections of plaques in the aortic arch of placebo- and stevioside-treated double
knockout (DKO) mice stained with oil red O (a) or with antibodies against macrophages (b) or oxidized-low-density lipoprotein (ox-LDL) (c). Total plaque volumes,
and macrophage, ox-LDL and lipid areas were lower in stevioside- (n¼14) than in placebo-treated (n¼20) DKO mice, the smooth muscle area and smooth muscle
cell-to-macrophage ratio of the plaque were increased (d). Data are means±s.e.m. *Po0.05, **Po0.01 and ***Po0.001 compared with placebo-treated DKO mice.
Stevioside on atherosclerosis in obesity
B Geeraert et al
573
International Journal of Obesity

As in adipose tissues, the expressions of Insr,Irs1, Irs2,
Glut4, Fabp4 and Lxrain the aorta of placebo-treated DKO
mice were lower than those in the aorta of lean control
C57BL6 mice (data not shown). Stevioside treatment only
increased the expression of Irs2. The expression of Irs2
correlated with Sod1 (Figure 3).
Furthermore, stevioside treatment decreased the expres-
sions of the chemotactic receptor Ccr2 and its ligand Ccl2,
important for the recruitment of monocytes/macrophages to
the vascular wall, supporting decreased inflammation. In
addition, it decreased the expression of the inflammatory
Nfkb1 and increased the expression of its inhibitor Nfkbia,
leading to a decreased Nfkb1/Nfkbiaratio. This ratio was
inversely related to Sod1 and Irs2, and positively to the ox-
LDL content of the plaque (Figure 4). In contrast, stevioside
treatment did not lower Icam1, Cd44,Cd68 and Cd36 and
did not increase Abca1 (data not shown).
Discussion
In normal physiology, adipose tissue stores energy in the
form of triglycerides, which can be broken down to release
free fatty acids. In obese individuals, impaired adipogenesis
is characterized by hypertrophic adipocytes, which are less
responsive to insulin and have a higher basal rate of fatty
acids release. This leads to increased fatty acids in the serum
and ectopic fat accumulation resulting in impaired insulin
signaling in non-adipose tissues.
24
We found that stevioside
treatment improved adipogenesis and glucose uptake in
visceral adipose tissue, evidenced by higher expressions of
Lxra, Fabp4 and Glut4.
25,26
The Glut4
27
and Fabp4
28
are
direct transcriptional targets for the LXR/retinoid X-receptor
heterodimer. Thus, the induction of Lxrain adipose tissue of
stevioside-treated mice supports the increased expression of
Glut4, which favors glucose uptake, and of Fabp4, which
improves fatty acid metabolism. The decrease in circulating
insulin could be explained by a lower need for insulin due to
improved insulin signaling, supported by the increased
expression of Irs1 and Irs2 in the adipose tissue of stevio-
side-treated mice. Irs1 is involved in the differentiation of
preadipocytes into adipocytes,
29
whereas Irs2 regulates the
insulin-induced glucose uptake.
30
The improved adipocyte
differentiation was associated with an increase in circulating
adiponectin, which correlated with Irs1 and Irs2.
We then investigated whether decreased hyperglycemia
and improved insulin sensitivity were associated with a
reduction of the obesity-induced oxidative stress. Previously,
it was demonstrated that the decreased expression of
antioxidant enzymes in adipose tissues of obese mice was
associated with increased ROS production.
31
Stevioside
Figure 3 Effect of stevioside on oxidative stress in the aortic arch. The RNA expressions of Sod1,Sod2, Sod3, Irs2 and Ppargwere lower in the aorta of double
knockout (DKO) mice than in lean mice. Stevioside increased all their expressions. All Sods were inversely related with the plaque oxidized-low-density lipoprotein
(ox-LDL) content. Sod1 correlated with Irs2 and adiponectin correlated with Pparg, Sod1 and Sod3. Ratios are means±s.e.m. *Po0.05, **Po0.01 and ***Po0.001
compared with lean control mice (C57BL6); and
$
Po0.05,
$$
Po0.01 and
$$$
Po0.001 compared with placebo DKO mice. Rs: Spearman’s correlation coefficients.
Stevioside on atherosclerosis in obesity
B Geeraert et al
574
International Journal of Obesity

treatment partially restored the expressions of Sod3 and Cat.
Both genes correlated strongly with adiponectin and Irs2.
In aggregate, we present a novel action of stevioside on
gene expression in adipose tissue associated with improved
adipogenesis, glucose uptake, insulin signaling and antiox-
idant defense. These observations prompted us to investigate
its effect on oxidative stress and insulin signaling in the aorta
in relation to atherosclerosis.
In the aortic arch, a reduction in macrophage, ox-LDL and
lipids were associated with the inhibition of atherosclerosis
in stevioside-treated mice. The decrease in macrophages was
likely due to reduced expressions of Ccl2 and its chemotactic
receptor Ccr2. In contrast, the expressions of Icam1 and
Cd44, which also mediate the interactions of monocytes
with other arterial cells, were not affected by stevioside
treatment.
The correlation between plaque macrophages and ox-LDL
is in agreement with macrophages being the principal
cellular source of ox-LDL.
17
Thus, the decrease in ox-LDL
could in part be due to a decrease in macrophages. In
addition, it can be explained by increased expression of ROS-
scavenging enzymes. As in the adipose tissue, we found
increased expression of the antioxidant Sods. Previously, we
showed that their induction, and the associated decrease in
ox-LDL, depends on the increase in Pparaactivity.
23
Their
increase in the vessel wall can also be due to the increase in
plasma adiponectin. Indeed, adiponectin correlated with
Sod,
32
and was inversely related to ox-LDL in diabetics.
33
The
subsequent decrease in ROS resulted in higher Irs2 expres-
sion, and thus improved insulin signaling.
34
In addition,
overexpression of adiponectin increased the Sods in mice.
35
Where there was ample evidence of diminished ROS and
ox-LDL generation by induction of anti-oxidant enzymes,
there was no evidence of enhanced ox-LDL clearance in
stevioside-treated mice. Indeed, the expressions of the
scavenger receptors CD36 and CD68 remained low.
Stevioside treatment also reduced the lipid content of the
plaques. This could be due to decreased accumulation of
lipids due to lowering of blood cholesterol and decreasing
accumulation of ox-LDL in the plaque. Most likely the
lowering of plaque lipids is not due to higher lipid efflux.
Indeed, the expression of Abca1 that is important for
inducing the reverse cholesterol transport, and thereby
reducing plaque lipids,
12
was not increased in stevioside-
treated mice. The latter could be due to the lack of effect on
LXRa. Indeed, in rosuvastatin-treated DKO mice an increase
in LXRawas associated with higher Abca1.
12
The lower macrophage, ox-LDL and lipid contents of the
atherosclerotic plaques in the aortic arch of DKO mice are
characteristic for a more stable plaque phenotype, which is
also supported by the higher SMC-to-macrophage ratio.
Interestingly, this higher ratio was not only due to a decrease
Figure 4 Effect of stevioside on inflammation in the aortic arch. Stevioside lowered the Nfkb1/Nfkbia ratio and the expressions of the proinflammatory Ccl2 and
Ccr2, which were elevated in the aorta of double knockout (DKO) mice. The Nfkb1/Nfkbia ratio inversely related to Sod1 and Irs2, and positively to the ox-LDL
content of the plaque. Ratios are means±s.e.m. *Po0.05, **Po0.01 and ***Po0.001 compared with lean control mice (C57BL6); and
$
Po0.05 and
$$
Po0.01
compared with placebo DKO mice. Rs: Spearman’s correlation coefficients.
Stevioside on atherosclerosis in obesity
B Geeraert et al
575
International Journal of Obesity

in macrophages, but also due to an increase in SMC. The
latter could be due to the decrease in ox-LDL. Indeed, ox-LDL
was found to induce metalloproteases, which degrade matrix
proteins, important for SMC migration, and, in addition, ox-
LDL induces SMC apoptosis.
36
It has been shown that Ppargexerts anti-inflammatory
actions by antagonizing the transcriptional activity of
Nfkb1.
37
Accordingly, stevioside-treated mice showed a lower
Nfkb1/Nfkbiaratio. Interestingly, the lower Nfkb1/Nfkbia
ratio in aortic extracts of stevioside-treated mice correlated
inversely with Sod1. The suppression of Nfkb1 could
partly be due to the suppression of O2

radicals by Sod1.
O2

radicals cause the phosphorylation and ubiquitination
of Nfkbiawith the subsequent translocation of Nfkb1 into
the nucleus and the transcription of proinflammatory
genes.
38
Thus, the reduction in oxidative stress may have
led to the reduction in inflammation, associated with
decreased monocyte infiltration, and thus a further decrease
in ROS and ox-LDL.
In conclusion, this is the first report showing an associa-
tion between stevioside treatment and increased adiponectin
and insulin sensitivity, improved antioxidant defense and
reduced atherosclerosis. The improved antioxidant defense
can be attributed mainly to increased expressions of Sods.
The latter correlated with decreased accumulation of ox-LDL
in circulation and the vessel wall. The decrease of ox-LDL by
stevioside is particularly important in view of our recent
observation that ox-LDL is associated with the metabolic
syndrome components obesity, hyperglycemia and insulin
resistance, and hyperlipidemia in the general population.
7,8
The decrease in ox-LDL was not due to an increase in PON1
in stevioside-treated mice.
A limitation of our study is that it is performed in mice.
However, all our data were obtained at a daily dosage that is
considered to be safe. Indeed, JECFA (Joint FAO/WHO Expert
Committee on Food Additives), an international food safety
organization that provides guidance on the safety of food
additives, has set a permanent Acceptable Daily Intake (ADI)
at 4 mg kg
1
body weight expressed as steviol, corresponding
to 11 mg kg
1
of stevioside (http://www.fao.org/ag/agn/agns/
jecfa_new_en.asp). The selected mice are hyperlipidemic and
hyperglycemic. Therefore, we do not know the effect of
stevioside in mice that are only hyperglycemic.
In conclusion, stevioside treatment was associated with
improved insulin signaling and antioxidant defense in both
the adipose tissue and the vascular wall of obese insulin-
resistant mice. The improved metabolism led to the inhibi-
tion of atherosclerotic plaque development and inducing
plaque stabilization. As yet, conflicting data about stevio-
side’s effects on insulin resistance and diabetes in humans
have been published.
39,40
However, the study groups were
very small. Moreover, patients were already treated with PPAR
agonists and/or statins, which are known to increase insulin
sensitivity. Thus, large-scale studies in humans are warranted.
Particularly, the action of stevioside for preventing insulin
resistance in obese persons requires further attention.
Conflict of interest
The authors declare no conflict of interest.
Acknowledgements
This study was supported in part by the Fonds voor
Wetenschappelijk Onderzoek–Vlaanderen (Program G.0548.08),
the OT/06/56 program and Interuniversity Attraction Poles
Program – Belgian Science Policy (P6/30). We thank Hilde
Bernar, Miche
`le Landeloos and Roxane Menten for excellent
technical assistance.
References
1 Geuns JM. Stevioside. Phytochemistry 2003; 64: 913–921.
2 Butler J, Rodondi N, Zhu Y, Figaro K, Fazio S, Vaughan DE et al.
Metabolic syndrome and the risk of cardiovascular disease in
older adults. J Am Coll Cardiol 2006; 47: 1595–1602.
3 Visser M, Bouter LM, McQuillan GM, Wener MH, Harris TB.
Elevated C-reactive protein levels in overweight and obese adults.
JAMA 1999; 282: 2131–2135.
4 Kelishadi R, Hashemi M, Mohammadifard N, Asgary S,
Khavarian N. Association of changes in oxidative and pro-
inflammatory states with changes in vascular function after
a lifestyle modification trial among obese children. Clin Chem
2008; 54: 147–153.
5 Van Guilder GP, Hoetzer GL, Greiner JJ, Stauffer BL, DeSouza CA.
Influence of metabolic syndrome on biomarkers of oxidative
stress and inflammation in obese adults. Obesity (Silver Spring)
2006; 14: 2127–2131.
6 Executive Summary of The Third Report of The National
Cholesterol Education Program (NCEP) Expert Panel on Detec-
tion, Evaluation, And Treatment of High Blood Cholesterol In
Adults (Adult Treatment Panel III). JAMA 2001; 285: 2486–2497.
7 Holvoet P, Kritchevsky SB, Tracy RP, Mertens A, Rubin SM, Butler J
et al. The metabolic syndrome, circulating oxidized LDL, and risk
of myocardial infarction in well-functioning elderly people in the
health, aging, and body composition cohort. Diabetes 2004; 53:
1068–1073.
8 Holvoet P, Lee DH, Steffes M, Gross M, Jacobs Jr DR. Association
between circulating oxidized low-density lipoprotein and inci-
dence of the metabolic syndrome. JAMA 2008; 299: 2287–2293.
9 Jeppesen PB, Gregersen S, Alstrup KK, Hermansen K. Stevioside
induces antihyperglycaemic, insulinotropic and glucagonostatic
effects in vivo: studies in the diabetic Goto-Kakizaki (GK) rats.
Phytomedicine 2002; 9: 9–14.
10 Jeppesen PB, Gregersen S, Poulsen CR, Hermansen K. Stevioside
acts directly on pancreatic beta cells to secrete insulin: actions
independent of cyclic adenosine monophosphate and adenosine
triphosphate-sensitive K+-channel activity. Metabolism 2000; 49:
208–214.
11 Verreth W, De Keyzer D, Pelat M, Verhamme P, Ganame J,
Bielicki JK et al. Weight loss-associated induction of peroxisome
proliferator-activated receptor-alpha and peroxisome prolifera-
tor-activated receptor-gamma correlate with reduced athero-
sclerosis and improved cardiovascular function in obese insulin-
resistant mice. Circulation 2004; 110: 3259–3269.
12 Verreth W, De Keyzer D, Davey PC, Geeraert B, Mertens A,
Herregods MC et al. Rosuvastatin restores superoxide dismutase
expression and inhibits accumulation of oxidized LDL in the
aortic arch of obese dyslipidemic mice. Br J Pharmacol 2007; 151:
347–355.
Stevioside on atherosclerosis in obesity
B Geeraert et al
576
International Journal of Obesity

13 Mertens A, Verhamme P, Bielicki JK, Phillips MC, Quarck R,
Verreth W et al. Increased low-density lipoprotein oxidation and
impaired high-density lipoprotein antioxidant defense are asso-
ciated with increased macrophage homing and atherosclerosis in
dyslipidemic obese mice: LCAT gene transfer decreases athero-
sclerosis. Circulation 2003; 107: 1640–1646.
14 Hasty AH, Shimano H, Osuga J, Namatame I, Takahashi A,
Yahagi N et al. Severe hypercholesterolemia, hypertriglyceridemia,
and atherosclerosis in mice lacking both leptin and the low
density lipoprotein receptor. JBiolChem2001; 276: 37402–37408.
15 Geuns JM, Buyse J, Vankeirsbilck A, Temme EH, Compernolle F,
Toppet S. Identification of steviol glucuronide in human urine.
J Agric Food Chem 2006; 54: 2794–2798.
16 Minne VJ, Compernolle F, Toppet S, Geuns JM. Steviol quanti-
fication at the picomole level by high-performance liquid
chromatography. J Agric Food Chem 2004; 52: 2445–2449.
17 Tsimikas S, Palinski W, Witztum JL. Circulating autoantibodies to
oxidized LDL correlate with arterial accumulation and depletion
of oxidized LDL in LDL receptor-deficient mice. Arterioscler
Thromb Vasc Biol 2001; 21: 95–100.
18 Mackness B, Quarck R, Verreth W, Mackness M, Holvoet P.
Human paraoxonase-1 overexpression inhibits atherosclerosis in
a mouse model of metabolic syndrome. Arterioscler Thromb Vasc
Biol 2006; 26: 1545–1550.
19 Ladunga I. Finding homologs to nucleotide sequences using network
BLAST searches. Curr Protoc Bioinformatics 2002. Chapter 3 Unit.
20 Ririe KM, Rasmussen RP, Wittwer CT. Product differentiation by
analysis of DNA melting curves during the polymerase chain
reaction. Anal Biochem 1997; 245: 154–160.
21 Holvoet P, Vanhaecke J, Janssens S, Van deWerf F, Collen D.
Oxidized LDL and malondialdehyde-modified LDL in patients
with acute coronary syndromes and stable coronary artery
disease. Circulation 1998; 98: 1487–1494.
22 Lin Y, Berg AH, Iyengar P, Lam TK, Giacca A, Combs TP et al.
The hyperglycemia-induced inflammatory response in adipo-
cytes: the role of reactive oxygen species. J Biol Chem 2005; 280:
4617–4626.
23 Desjardins F, Sekkali B, Verreth W, Pelat M, De KD, Mertens A
et al. Rosuvastatin increases vascular endothelial PPARgamma
expression and corrects blood pressure variability in obese
dyslipidaemic mice. Eur Heart J 2008; 29: 128–137.
24 Boden G, Chen X, Ruiz J, Heifets M, Morris M, Badosa F. Insulin
receptor down-regulation and impaired antilipolytic action of
insulin in diabetic patients after pancreas/kidney transplanta-
tion. J Clin Endocrinol Metab 1994; 78: 657–663.
25 Fernyhough ME, Okine E, Hausman G, Vierck JL, Dodson MV.
PPARgamma and GLUT-4 expression as developmental regula-
tors/markers for preadipocyte differentiation into an adipocyte.
Domest Anim Endocrinol 2007; 33: 367–378.
26 Hummasti S, Laffitte BA, Watson MA, Galardi C, Chao LC,
Ramamurthy L et al. Liver X receptors are regulators of adipocyte
gene expression but not differentiation: identification of apoD as
a direct target. J Lipid Res 2004; 45: 616–625.
27 Laffitte BA, Chao LC, Li J, Walczak R, Hummasti S, Joseph SB et al.
Activation of liver X receptor improves glucose tolerance through
coordinate regulation of glucose metabolism in liver and adipose
tissue. Proc Natl Acad Sci USA 2003; 100: 5419–5424.
28 Liu QY, Quinet E, Nambi P. Adipocyte fatty acid-binding
protein (aP2), a newly identified LXR target gene, is induced by
LXR agonists in human THP-1 cells. Mol Cell Biochem 2007; 302:
203–213.
29 Fasshauer M, Klein J, Kriauciunas KM, Ueki K, Benito M,
Kahn CR. Essential role of insulin receptor substrate 1 in
differentiation of brown adipocytes. Mol Cell Biol 2001; 21: 319–329.
30 Fasshauer M, Klein J, Ueki K, Kriauciunas KM, Benito M,
White MF et al. Essential role of insulin receptor substrate-2 in
insulin stimulation of Glut4 translocation and glucose uptake in
brown adipocytes. J Biol Chem 2000; 275: 25494–25501.
31 Furukawa S, Fujita T, Shimabukuro M, Iwaki M, Yamada Y,
Nakajima Y et al. Increased oxidative stress in obesity and
its impact on metabolic syndrome. J Clin Invest 2004; 114:
1752–1761.
32 Adachi T, Inoue M, Hara H, Maehata E, Suzuki S. Relationship of
plasma extracellular-superoxide dismutase level with insulin
resistance in type 2 diabetic patients. J Endocrinol 2004; 181:
413–417.
33 Prior SL, Gable DR, Cooper JA, Bain SC, Hurel SJ, Humphries SE
et al. Association between the adiponectin promoter rs266729
gene variant and oxidative stress in patients with diabetes
mellitus. Eur Heart J 2009; 30: 1263–1269.
34 Taniyama Y, Hitomi H, Shah A, Alexander RW, Griendling KK.
Mechanisms of reactive oxygen species-dependent downregula-
tion of insulin receptor substrate-1 by angiotensin II. Arterioscler
Thromb Vasc Biol 2005; 25: 1142–1147.
35 Fujita K, Maeda N, Sonoda M, Ohashi K, Hibuse T, Nishizawa H
et al. Adiponectin protects against angiotensin II-induced cardiac
fibrosis through activation of PPAR-alpha. Arterioscler Thromb
Vasc Biol 2008; 28: 863–870.
36 Mertens A, Holvoet P. Oxidized LDL and HDL: antagonists in
atherothrombosis. FAS EB J 2001; 15: 2073–2084.
37 Ajuwon KM, Spurlock ME. Adiponectin inhibits LPS-induced
NF-kappaB activation and IL-6 production and increases PPAR-
gamma2 expression in adipocytes. Am J Physiol Regul Integr Comp
Physiol 2005; 288: R1220–R1225.
38 Barnes PJ, Karin M. Nuclear factor-kappaB: a pivotal transcription
factor in chronic inflammatory diseases. N Engl J Med 1997; 336:
1066–1071.
39 Barriocanal LA, Palacios M, Benitez G, Benitez S, Jimenez JT,
Jimenez N et al. Apparent lack of pharmacological effect of steviol
glycosides used as sweeteners in humans. A pilot study of
repeated exposures in some normotensive and hypotensive
individuals and in Type 1 and Type 2 diabetics. Regul Toxicol
Pharmacol 2008; 51: 37–41.
40 Gregersen S, Jeppesen PB, Holst JJ, Hermansen K. Antihypergly-
cemic effects of stevioside in type 2 diabetic subjects. Metabolism
2004; 53: 73–76.
Stevioside on atherosclerosis in obesity
B Geeraert et al
577
International Journal of Obesity