Ajoene, A stable garlic by-product, inhibits high fat diet-induced hepatic steatosis and oxidative injury through LKB1-dependent AMPK activation

Article abstract of DOI:10.1089/ars.2010.3190

Hepatic steatosis, a hepatic component of metabolic syndrome, is common and may progress to steatohepatitis and cirrhosis. The liver X receptor-α (LXRα)-sterol regulatory element binding protein-1c (SREBP-1c) pathway plays a key role in hepatic steatosis. This study investigated the potential of ajoene, a stable garlic by-product, to inhibit high fat diet (HFD)-induced hepatic steatosis and the underlying mechanism. Ajoene treatment attenuated fat accumulation and induction of lipogenic genes in the liver of HFD-fed mice. Blood biochemical analyses and histopathologic examinations showed that ajoene prevented liver injury with the inhibition of oxidative stress, as evidenced by thiobarbituric acid reactive substances formation and nitrotyrosinylation. Moreover, ajoene treatment inhibited LXRα agonist (T0901317)-mediated SREBP-1c activation, and transactivation of the lipogenic target genes in hepatocytes. Ajoene was found to activate AMP-activated protein kinase (AMPK) via LKB1, responsible for the inhibition of p70 ribosomal S6 kinase-1 (S6K1). The ability of ajoene to repress T0901317-induced SREBP-1c expression was antagonized by inhibition of AMPK or activation of S6K1, supporting the role of these kinases in the antisteatotic effect. Our results demonstrate that ajoene has an effect of activating AMPK through LKB1 and inhibit S6K1 activity, contributing to the prevention of SREBP-1c-mediated hepatic lipogenesis via the inhibition of LXRα activity.

Full text of DOI:10.1089/ars.2010.3190

ANTIOXIDANTS & REDOX SIGNALING
Volume 14, Number 2, 2011
ª Mary Ann Liebert, Inc.
ORIGINAL RESEARCH COMMUNICATION
DOI: 10.1089=ars.2010.3190
Ajoene, A Stable Garlic By-Product, Inhibits High
Fat Diet-Induced Hepatic Steatosis and Oxidative Injury
Through LKB1-Dependent AMPK Activation
1
1
1
1
2
Chang Yeob Han, Sung Hwan Ki, Young Woo Kim, Kyoung Noh, Da Yeon Lee, Bomi Kang,
2
2
Jae-Ha Ryu, Raok Jeon, Eun Hyun Kim, Se Jin Hwang, and Sang Geon Kim
2
3
3
1
Abstract
Hepatic steatosis, a hepatic component of metabolic syndrome, is common and may progress to steatohepatitis
and cirrhosis. The liver X receptor-a (LXRa)-sterol regulatory element binding protein-1c (SREBP-1c) pathway
plays a key role in hepatic steatosis. This study investigated the potential of ajoene, a stable garlic by-product, to
inhibit high fat diet (HFD)-induced hepatic steatosis and the underlying mechanism. Ajoene treatment attenu-
ated fat accumulation and induction of lipogenic genes in the liver of HFD-fed mice. Blood biochemical analyses
and histopathologic examinations showed that ajoene prevented liver injury with the inhibition of oxidative
stress, as evidenced by thiobarbituric acid reactive substances formation and nitrotyrosinylation. Moreover,
ajoene treatment inhibited LXRa agonist (T0901317)-mediated SREBP-1c activation, and transactivation of the
lipogenic target genes in hepatocytes. Ajoene was found to activate AMP-activated protein kinase (AMPK) via
LKB1, responsible for the inhibition of p70 ribosomal S6 kinase-1 (S6K1). The ability of ajoene to repress
T0901317-induced SREBP-1c expression was antagonized by inhibition of AMPK or activation of S6K1, sup-
porting the role of these kinases in the antisteatotic effect. Our results demonstrate that ajoene has an effect of
activating AMPK through LKB1 and inhibit S6K1 activity, contributing to the prevention of SREBP-1c-mediated
hepatic lipogenesis via the inhibition of LXRa activity. Antioxid Redox Signal. 14, 187–202.
Introduction
gene of LXRa, and is a key transcriptional regulator of lipo-
genic genes encoding enzymes to promote fat accumulation in
onalcoholic fatty liver disease (NAFLD) is a com- the liver, which include fatty acid synthase (FAS), acetyl-CoA
N
mon liver disease that refers to the range of liver damage carboxylase (ACC), and stearoyl-CoA desaturase-1 (SCD-1)
from steatosis to nonalcoholic steatohepatitis and cirrhosis (33). Chemical activation of LXRa promotes SREBP-1c-
35). Moreover, NAFLD is considered to be the major hepatic dependent induction of the lipogenic genes, causing increases
(
component of metabolic syndrome (25). Features of metabolic in fatty acid synthesis and further progression to hepatic
syndrome include obesity, insulin resistance, and cardiovas- steatosis and hypertriglyceridemia (39). Therefore, the lipo-
cular diseases. In obese people, most with insulin resistance, genic pathway mediated by LXRa-SREBP-1c is an attractive
excessive fat is accumulated in the liver and the elevated he- target for the prevention and=or treatment of hepatic steatosis
patic lipid content is closely associated with pathogenesis of and steatohepatitis.
the syndrome (25). In hepatocytes, fat amounts are enhanced
AMP-activated protein kinase (AMPK) is an intracellular
by induction of lipogenesis, as well as by increased delivery of sensor for energy homeostasis (30). Diverse physiologic re-
fatty acids and diminished export of triglyceride (TG), which sponses associated with lipid and glucose metabolism are
eventually results in chronic liver diseases in conjunction with mediated by the AMPK-dependent pathway. The activation
systemic metabolic dysfunction (25, 35).
of this kinase inhibits hepatic lipogenesis mainly through in-
Liver X receptor-a (LXRa) is a nuclear receptor that acts as a hibitory phosphorylation of ACC, a rate-controlling en-
regulator to increase lipid synthesis in the liver (39). Sterol zyme of fatty acid synthesis. Cholesterol production is also
regulatory element binding protein-1c (SREBP-1c) is a target inhibited by AMPK activation in the liver via suppression of
1
College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Seoul, Korea.
2
College of Pharmacy, Sookmyung Women’s University, Seoul, Korea.
3
College of Medicine, Hanyang University, Seoul, Korea.
1
87

188
HAN ET AL.
3
-hydroxy-3-methylglutaryl-CoA reductase (48). In addition, by regioselective radical addition of thiolacetic acid, to make
AMPK activation reduces gluconeogenesis in liver and en- vinyl thioacetate. Deprotection of thioacetate with hydrox-
hances glucose uptake in skeletal muscle (30). Activated mu- ide in methanol and subsequent sulfenylation of the en-
tant of AMPK inhibits mammalian target of rapamycin=p70 ethiolate with S-allyl p-toluenesulfonylthioate yielded vinyl
ribosomal S6 kinase-1 (S6K1) signaling cascade (20). Our disulfide. The resultant disulfide was chemoselectively oxi-
previous finding indicates that AMPK activation and AMPK- dized by treatment of meta-chloroperoxybenzoic acid
dependent S6K1 inhibition contributes to antidiabetic and (m-CPBA) (Fig. 1A). The chemical structure of (Z)-ajoene
antisteatotic action (2, 18). Furthermore, these enzymes op- was confirmed by mass spectrometry and NMR analyses
positely regulate the LXRa activity via phosphorylation at (Figs. 1B and 1C).
different residues, which affects LXRa-mediated steatosis in
the liver (18).
Animal treatment
Garlic (Allium sativum) is a widely consumed natural
product and has medicinal properties. Among the constitu-
ents of garlic, ajoene (4,5,9-trithiadodeca-1,6,11-triene-9-
oxide) is a biologically active organosulfur compound (19),
and exerts beneficial effects such as anticancer, anti-athero-
sclerosis, and antimicrobial actions (1, 37, 46). Allicin also has
biological actions similar to ajoene, but is highly unstable as
shown by thermostability analysis (12). Ajoene is largely
generated from allicin in crushed garlic by sulfoxide re-
arrangement in heat condition (12), and thus has chemical
stability (15). Moreover, ajoene can be easily made by organic
synthesis, which may offer an advantage for development of a
drug candidate and clinical use. Regardless of the intensive
studies of garlic products, molecular mechanisms mediated
by specific garlic component(s) are poorly understood; fur-
thermore, the effects of ajoene on lipid metabolism and its
associated conditions or diseases have never been explored.
This study investigated whether ajoene inhibits high fat
diet (HFD)-induced hepatic steatosis and, if so, the mecha-
nistic basis. Our findings demonstrate that ajoene has the
ability to prevent SREBP-1c-dependent lipogenesis in the liver
via the inhibition of LXRa activity and thereby reverse hepatic
steatosis and steatohepatitis. Moreover, we found that ajoene
All animal experiments were approved by the Seoul Na-
tional University Animal Experiment Ethics Committee, and
were conducted in accordance with the institutional guide-
lines for care and use of laboratory animals. Male C57BL=6
mice were obtained from Charles River Orient (Seoul, Korea)
and were acclimatized for 1 week in a clean room at the An-
imal Center for Pharmaceutical Research, Seoul National
University. At 7 weeks of age, C57BL=6 mice were started on
either a normal diet (ND) or an HFD (60% of kcal as fat,
Product #D12492, Research Diets, New Brunswick, NJ) for 8
weeks: the animals were distributed into three treatment
groups [i.e., HFDþvehicle, HFDþajoene (10 mg=kg=day), and
HFDþajoene (30 mg=kg=day), n ¼ 8 each] after 4 weeks of diet
feeding. Then, ajoene dissolved in 40% polyethyleneglycol
#
400 was orally administered to mice 5 times per week during
the last 4 weeks of diet feeding. Six hours after last ajoene
administration, mice were fasted overnight (*16 h), and
blood samples were drawn by retro-orbital puncture. The
animals were sacrificed by carbon dioxide euthanasia.
Hematoxylin and eosin, and Oil Red O stainings
The left lateral lobe of the liver was sliced, and tissue slices
treatment activates AMPK via liver kinase B1 (LKB1) and were fixed in 10% buffered-neutral formalin for 6 h. The fixed
inhibits S6K1 in hepatocytes, and that the beneficial anti- liver tissue slices were embedded in paraffin, and cut into
steatotic effect of ajoene is mediated with the regulation of 4 mm thick sections, which were stained with hematoxylin and
AMPK-S6K1 pathway, providing the possibility of a new eosin (H&E) (27). Histopathology was assessed by light mi-
therapeutic application of this agent for NAFLD and=or other croscopy. A certified pathologist examined the samples in a
metabolic dysfunctions.
blinded fashion. For Oil Red O (Sigma) staining, 4 mm sections
were cut from frozen optimal cutting temperature samples,
affixed to microscope slides, and allowed to air-dry overnight
at room temperature. The liver sections were stained in fresh
Oil Red O for 10 min and rinsed in water (18).
Materials and Methods
Reagents
Antibodies directed against SREBP-1, AMPK, and LKB1
were supplied by Santa Cruz Biotechnology (Santa Cruz, CA).
Antiphospho-AMPK, antiphospho-ACC, anti-ACC, antipho-
spho-S6, anti-S6, and antiphospho-LKB1 antibodies were
obtained from Cell Signaling (Beverly, MA). Antibodies rec-
ognizing inducible nitric oxide synthase (iNOS) and cy-
clooxygenase-2 (COX-2) were provided from BD biosciences
Triglyceride measurement
Samples of mouse liver (0.3 g) were homogenized in 0.1 M
Tris–acetate buffer (pH 7.4) containing 0.1 M potassium
chloride and 1 mM EDTA. Six volumes of chloroform:meth-
anol (2:1) were then added. After vigorous stirring, the mix-
tures were incubated on ice for 1 h and then centrifuged at 800
g for 3 min. The resulting lower phase was aspirated. The TG
content was determined using Sigma Diagnostic Triglyceride
Reagents (18).
(San Jose, CA) and Cayman (Ann Arbor, MI), respectively.
Horseradish peroxidase-conjugated goat anti-rabbit and goat
anti-mouse IgGs were purchased from Zymed Laboratories
(San Francisco, CA). 22(R)-Hydroxycholesterol, GW3965,
diallyl sulfide (DAS), diallyl disulfide (DADS), and other re-
agents were supplied from Sigma-Aldrich (St. Louis, MO).
Blood chemistry
Plasma alanine aminotransferase (ALT), TG, and total
cholesterol were analyzed using Spectrumꢀ, an automatic
blood chemistry analyzer (Abbott Laboratories, Abbott Park,
Chemical synthesis and characterization of ajoene
Ajoene was synthesized according to the modified Hun- IL). Fasting glucose in plasma was measured using Accu-
ter’s method (17). Allyl thiol was propargylated, followed Chek Active (Roche, Germany).

INHIBITION OF HEPATIC STEATOSIS BY AJOENE
189
FIG. 1. Chemical synthesis and characterization of ajoene. (A) The synthetic pathway of (Z)-ajoene. (Z)-Ajoene was
synthesized as described in the Materials and Methods section. (B) Electron impact (EI) mass spectrum. (C) NMR analyses.
Hydrogen-1 NMR (upper) and carbon-13 NMR (lower).
Thiobarbituric acid reactive substances assay
against nitrotyrosine (Millipore, Billerica, MA). Briefly, tissue
sections were deparaffinized and were incubated with rabbit
anti-nitrotyrosine (1:200) for 16 h, followed by incubation
with Cy3-conjugated donkey anti-rabbit antibody for 2 h (27).
Thiobarbituric acid reactive substances (TBARS) formation
was determined in liver homogenates, and the amounts were
referenced as moles of malonyldialdehyde (MDA) using an
MDA standard (4). Briefly, MDA-thiobarbituric acid adduct
was formed by the reaction of MDA present in the sample and Determination of reduced glutathione
thiobarbituric acid solution under high temperature (908–
Reduced glutathione (GSH) in the liver tissue was quanti-
fied using a commercially available GSH determination kit
1
008C) and acidic conditions. The absorbance was measured
at 532 nm.
(Oxis International, Portland, OR). Briefly, the GSH-400
method was based on a chemical reaction which proceeded in
two steps. The first step led to the formation of substitution
Immunohistochemistry
Liver specimens were fixed in 10% formalin, embedded in products (thioethers) between 4-chloro-1-methyl-7-trifluromethyl-
paraffin, cut into 4-mm-thick sections, and mounted on slides. quinolinum methylsulfate and all mercaptans present in the
Tissue sections were immunostained with antibodies directed samples. The second step included b-elimination reaction

190
HAN ET AL.
0
0
under an alkaline condition. This reaction was mediated by CTGTCCAG-3 (antisense); human SCD-1, 5 -CCTCTA
0
0
3
0% NaOH which specifically transformed the substituted CTTGGAAGACGACATTCGC-3 (sense) and 5 -GCAGCC
0
product (GSH-thioether) into a chromophoric thione.
GAGCTTTGTAAGAGCGGT-3 (antisense); human GAPDH,
-GAAGATGGTGATGGGATTTC-3 (sense) and 5 -GAAG
GTGAAGGTCGGAGTC-3 (antisense); mouse LXRa, 5 -
TGCCATCAGCATCTTCTCTG-3 (sense) and 5 -GGCTCAC
CAGCTTCATTAGC-3 (antisense); mouse SREBP-1c, 5 -AAC
GTCACTTCCAGCTAGAC-3 (sense) and 5 -CCACTAAGG
TGCCTACAGAGC-3 (antisense); mouse FAS, 5 -AGCGGC
0
0
0
5
0
0
Determination of glutathione peroxidase
and catalase activities
0
0
0
0
The activities of glutathione peroxidase (GPx) and catalase
were quantified using commercially available systems, GPx
0
0
0
0
(Cayman, Ann Arbor, MI) and catalase assay kits (Sigma-
0
0
CATTTCCATTGCCC-3 (sense) and 5 -CCATGCCCAGAG
Aldrich), respectively. Briefly, GPx activity was measured by
a coupled reaction with glutathione reductase. Oxidized GSH
0
0
GGTGGTTG-3 (antisense); mouse ACC1, 5 -GTCAGCGGA
0
0
TGGGCGGAATG-3 (sense) and 5 -CGCCGGATGCCATG
CTCAAC-3 (antisense); mouse SCD-1, 5 -CCGGAGACC
CTTAGATCGA-3 (sense) and 5 -TAGCCTGTAAAAGAT
TTCTGCAAACC-3 (antisense); mouse carnitine palmitoyl
transferase-1 (CPT-1), 5 -GTCGCTTCTTCAAGGTCTGG-3
sense) and 5 -TTCTTTCGTCGTGCAAGCTA-3 (antisense);
mouse peroxisome proliferator-activated receptor g coacti-
vator-1a (PGC-1a), 5 -ACGAGGCCAGTCCTTCCTCC-3
sense) and 5 -AGCTCTGAGCAGGGACGTCT-3 (antisense);
mouse tumor necrosis factor-a (TNFa), 5 -TCCCAGGTT
CTCTTCAAGGGA-3 (sense) and 5 -GGTGAGGAGCACG
TAGTCGG-3 (antisense); and mouse GAPDH, 5 -AACGACC
CCTTCATTGAC-3 (sense) and 5 -TCCACGACATACTCAG
CAC-3 (antisense).
(GSSG), produced upon reduction of hydroperoxide by GPx,
0
0
was recycled to its reduced state by glutathione reductase and
NADPH (36). The oxidation of NADPH to NADP was ac-
0
0
þ
0
companied by a decrease in absorbance at 340 nm. Catalase
activity was determined by measuring hydrogen peroxide
substrate remaining after the action of catalase. First, catalase
converted hydrogen peroxide to water and oxygen, which
was stopped with sodium azide. An aliquot of the reaction
mixture was colorimetrically assayed using a substituted
phenol (3,5-dichloro-2-hydroxybenzene-sulfonic acid), which
coupled to 4-aminoantipyrine in the presence of hydro-
gen peroxide and horseradish peroxidase to yield a red
quinoneimine dye (N-(4-antipyryl)-3-chloro-5-sulfonate-p-
benzoquinone-monoimine) (absorbance at 520 nm).
0
0
0
0
(
0
0
0
0
(
0
0
0
0
0
0
0
0
Transient transfection and reporter gene assays
Cell culture and in vitro assays
The cells were transiently transfected with pGL2-FAS (1 mg)
or TK-CYP7A1-LXREx3-LUC (0.5 mg) for 3 h in the presence of
Lipofectamineꢀ (Invitrogen, Carlsbad, CA) reagent. The ac-
tivity of luciferase was measured by adding luciferase assay
reagent (Promega, Madison, WI). The FAS reporter plasmid
HepG2 cells, a human hepatocyte-derived cell line, were
purchased from ATCC (Rockville, MD). Primary hepatocytes
were isolated from male Sprague-Dawley rats (23). Cells were
5
plated at 1ꢀ10 per well in 6-well plates, and wells with 70%–
8
0% confluency were used. All cells were maintained in
(pGL2-FAS) was a gift from Dr. T. F. Osborne (University of
Dulbecco’s modified Eagle’s medium containing 10% fetal
California, Irvine, CA). Dr. D. J. Mangelsdorf (Howard Hughes
Medical Institute, University of Texas Southwestern Medical
Center, Dallas, TX) provided the luciferase reporter plasmid
TK-CYP7a-LXRE(X3)-LUC that contains three tandem copies
of the sequence (5 -gcttTGGTCActcaAGTTCAagtta-3 ) from
the rat CYP7A1 gene.
bovine serum, 50 units=ml penicillin and 50 mg=ml strepto-
mycin at 378C in humidified atmosphere with 5% CO
cells were treated with vehicle or LXR agonist [i.e., T0901317,
0 mM GW3965, and 10 mM 22(R)-hydroxycholesterol] in the
presence or absence of ajoene for 12 h. In addition, the effect of
0 mM DAS or DADS on ACC phosphorylation was compared
2
. The
1
0
0
3
with that of ajoene.
Chromatin immunoprecipitation assay
RNA isolation and real-time PCR assays
The cells were treated with T090 or T090þajoene for 6 h fol-
Total RNA was extracted using Trizol (Invitrogen, Carls- lowing dominant negative (DN)-AMPK transfection, and then
bad, CA) according to the manufacturer’s instructions. Total formaldehyde was added to the cells to a final concentration of
RNA (1 mg) was reverse-transcribed using an oligo(dT)16 pri- 1% for cross-linking of chromatin. The chromatin immuno-
mer to obtain cDNA. The cDNA was amplified by polymerase precipitation assay was performed according to the chromatin
chain reaction (PCR). Real-time PCR was performed with a immunoprecipitation (ChIP) assay kit protocol (Upstate
Light Cycler 1.5 apparatus (Roche, Mannheim, Germany) Biotechnology, Lake Placid, NY). PCR was performed using
using a Light Cycler DNA master SYBR green-I kit according the specific primer flanking the LXRE region of the SREBP-1c
to the manufacturer’s instructions. The levels of target gene promoter (sense, 5’-TCAGGGTGCCAGCGAACC-3’, and
mRNAs were normalized to those of glyceraldehyde-3- antisense, 5’-GCTCGAGTTTCACCCCGC-3’, 207 bp).
phosphate dehydrogenase (GAPDH) mRNA. The following
primer sequences were used: human SREBP-1c, 5’-CGA
Immunoblot analysis
CATCGAAGACATGCTTCAG-3’ (sense) and 5’-GGAAGGC
TTCAAGAGAGGAGC-3’ (antisense); human LXRa, 5’-GAT
Total cell lysates were prepared as previously described
CGAGGTGATGCTTCTGGAG-3’ (sense) and 5’-CCCTGCTT (18). Briefly, the cells were centrifuged at 3000 g for 3 min and
TGGCAAAGTCTTC-3’ (antisense); human FAS, 5’-GACATC allowed to expand osmotically to the point of lysis after the
GTCCATTCGTTTGTG-3’ (sense) and 5’-CGGATCACCTTCT addition of lysis buffer. Lysates were centrifuged at 10,000 g
0
TGAGCTCC-3’ (antisense); human ACC1, 5 -GCTGCTCG for 10 min to obtain supernatants and were stored at -708C
0
0
GATCACTAGTGAA-3 (sense) and 5 -TTCTGCTATCAGT until use. Immunoblot analysis was performed according to

INHIBITION OF HEPATIC STEATOSIS BY AJOENE
191
previously published procedures (18). Protein bands of in- by ajoene treatment (Fig. 3D). Overall, our results demon-
terest were developed using an ECL chemiluminescence strate that ajoene has the effect to attenuate hepatic steatosis
system (Amersham, Buckinghamshire, UK). Equal protein and protect the liver from injury in the progression of steatosis
loading was verified by immunoblotting for actin.
to steatohepatitis.
Data analysis
Inhibition of HFD-induced oxidative stress
One-way analyses of variance procedures were used to
assess significant differences among treatment groups. For
each significant treatment effect, the Newman–Keuls test was
utilized to compare multiple group means. All statistical an-
alyses were performed using SPSS for Windows (version 15.0;
SPSS, Chicago, IL). The criterion for statistical significance
was set at p < 0.05 or 0.01.
To test whether ajoene has a protective effect on HFD-
induced oxidative stress, formation of hepatic TBARS, an in-
dex of lipid peroxidation, was examined. As expected, HFD
feeding increased TBARS formation in the liver tissue, which
was inhibited by ajoene treatment (Fig. 4A). Moreover, the
beneficial effect of ajoene on HFD-induced free radical stress
was confirmed by immunohistochemical analyses for ni-
trotyrosine, a representative marker of protein oxidation
mediated by reactive nitrogen species (RNS) (44). The inten-
sity of nitrotyrosinylation increased in the liver tissue of mice
fed an HFD, which was more severe around the pericentral
area (Fig. 4B). Apparently, ajoene treatment inhibited oxida-
tive stress-induced pathologic alteration. As expected, HFD
feeding caused a decrease in hepatic GSH content, which was
also antagonized by ajoene treatment (Fig. 4C). Oxidative
injury may result from a decrease in the capacity of cellular
antioxidant enzymes. Ajoene treatment prevented the ability
of HFD-feeding to diminish GPx and catalase activities (Fig.
Results
Inhibition of HFD-induced hepatic steatosis
To investigate an inhibitory effect of ajoene on fatty liver
in vivo, mice fed an HFD (60% fat) were treated with vehicle or
ajoene. Body weight gain was noticeably increased in mice fed
an HFD for 8 weeks (Fig. 2A). Treatment of mice with
30 mg=kg ajoene per day for the final 4 weeks during 8 weeks
of HFD feeding significantly reduced the rate of body weight
gain. However, treatment of ajoene did not change the
amounts of food intake (Fig. 2B). In addition, ajoene treatment
at the dose of 30 mg=kg=day significantly inhibited the ability
of HFD to increase the liver weight (Fig. 2C).
Since fat accumulation in the liver is the main histological
feature of NAFLD, hepatic lipid accumulation was measured
in the mice. As shown in Figure 2D, Oil red O staining con-
firmed the remarkable increase in hepatic fat in mice fed an
HFD for 8 weeks, which was notably diminished by treatment
of ajoene. The content of hepatic TG, the most abundant lipid
accumulated in hepatocytes in NAFLD, also successfully in-
creased in the liver of mice fed an HFD. The accumulation of
hepatic TG was significantly decreased by ajoene treatment at
the daily dose of either 10 mg=kg or 30 mg=kg (Fig. 2D).
4
D). These results show that fat accumulation causes oxida-
tive stress and increases hepatocyte injury (i.e., TBARS for-
mation and protein nitration in the liver), and which was
inhibited by ajoene as a result of the increase in antioxidant
capacity.
Inhibition of lipogenic gene induction
Given the role of LXRa-SREBP-1c pathway in hepatic fat
accumulation, we examined the effects of ajoene on the ex-
pression of these genes in the liver of mice fed an HFD. The
increases in mRNA levels of LXRa and SREBP-1c were ob-
served in the HFD-fed mice, whereas treatment with ajoene
inhibited HFD feeding-induced LXRa and SREBP-1c gene
induction, as assessed by quantitative real-time PCR assays
Protection of HFD-induced liver injury
(
Fig. 5A). Consistently, the induction of downstream lipo-
Because excessive fat accumulation in the liver leads to
hepatitis, we determined the effect of ajoene treatment on
HFD-induced liver injury. The blood biochemical analysis
showed that ALT activity in plasma was increased in HFD-fed
mice, which was prevented by treatment with ajoene at the
dose of 30 mg=kg=day for the final 4 weeks (Fig. 3A). Con-
sistently, HFD feeding for 8 weeks resulted in the formation of
large cytoplasmic lipid droplets, hepatocyte death, and infil-
tration of inflammatory cells in the liver, as assessed by his-
topathologic examinations (Fig. 3B). Numerous vacuoles and
degeneration of the hepatocytes were observed in the peri-
central and intermediate zones of the liver in HFD-fed mice.
The inclusion of lipid droplets and eosinophilic degeneration
were markedly reduced by a low dose (10 mg=kg=day) of
ajoene treatment. In particular, hepatocytes were intact in all
areas and virtually no vacuolar abnormality was observed in
the liver tissue of HFD-fed mice treated with 30 mg=kg=day of
ajoene (Fig. 3B). In addition, the levels of inflammatory gene
products were examined in the liver samples. TNFa mRNA
level was increased in HFD-fed mice, which was inhibited by
ajoene treatment (Fig. 3C). HFD feeding also induced iNOS
genic genes including FAS, ACC, and SCD-1 was decreased
by ajoene treatment (Fig. 5B), indicating that repression of the
LXRa-mediated lipogenic process contributes to ajoene’s in-
hibition of hepatic fat accumulation.
In addition, the levels of TG and total cholesterol in plasma
were diminished by ajoene in mice fed an HFD (Fig. 5C),
being consistent with its inhibition of liver TG accumulation
and lipogenic gene induction. Insulin resistance is often ac-
companied by the condition of fatty liver; it also occurs in an
animal model fed on an HFD (51). In addition to the decrease
in hepatic fat accumulation, ajoene treatment decreased HFD-
induced fasting glucose and insulin contents in plasma (Fig.
4
D). Taken together, our results demonstrate that ajoene has
the effect to inhibit hepatic steatosis as a consequence of the
inhibition of lipogenic gene induction, and that this anti-
steatotic efficacy is accompanied by the improvement in
glucose metabolism.
Inhibition of T090-induced LXRa-SREBP-1c activation
To verify the inhibitory action of ajoene on LXRa-dependent
and COX-2 in the liver of HFD-fed mice, which was reversed SREBP-1c activation, the effect of ajoene on the activation of

192
HAN ET AL.
FIG. 2. Antisteatotic effect of ajoene in mice. (A) The effect of ajoene on body weight gain in mice. Male C57BL=6 mice
were fed on either a normal diet (ND) or high-fat diet (HFD) for 8 weeks. During the final 4 weeks of diet feeding, ajoene (10
or 30 mg=kg=day) was orally administered to mice 5 times per week (n ¼ 8 animals per each treatment group). (B) Food
intake. (C) Changes in the liver weight. (D) Antisteatotic effect of ajoene. The extent of hepatic TG accumulation was assessed
in HFD-fed mice treated with either vehicle or ajoene. The left panel shows Oil Red O staining of the liver (100X). Statistical
#
##
significance of differences between each treatment group and ND (**p < 0.01) or HFD alone ( p < 0.05, p < 0.01) was
determined. Ajo, ajoene; N.S., not significant. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article at www.liebertonline.com=ars).
LXRa by T0901317 (T090, an LXRa agonist) was examined in represses transactivation of the LXRa gene. The activity of
an in vitro cell model. Since the promoter region of LXRa gene CYP7A1-LXRE-luciferase (CYP7A1) gene was enhanced by
contains an LXR response element (LXRE), LXRa is induced by treatment of the cells with T090, which was prevented by
its own regulatory induction mechanism (29). In subsequent ajoene treatment (Fig. 6A, right).
experiments, the expression of LXRa gene was monitored us-
Since LXRa is critical for SREBP-1c induction and the del-
ing real-time PCR assay in HepG2 cells. As expected, the LXRa eterious effects of LXRa activation mainly result from the in-
mRNA level was increased by treatment with T090 for 12 h, crease in SREBP-1c in hepatocytes, the effect of ajoene on
whereas this effect was significantly abrogated by simulta- T090-mediated SREBP-1c expression was also measured. The
neous ajoene treatment at either 10 mM or 30 mM concentration level of SREBP-1c protein was induced by T090 treatment for
(Fig. 6A, left). Next, we determined whether ajoene treatment 12 h in both HepG2 cells and primary rat hepatocytes (Fig.

INHIBITION OF HEPATIC STEATOSIS BY AJOENE
193
FIG. 3. Inhibition of HFD-
induced liver injury. (A) Plasma
ALT activity. (B) H&E staining in
the liver of ND-fed mice (ND) or
HFD-fed mice (HFD) treated with
vehicle or 10 mg=kg=day (Low) or
30 mg=kg=day (High) ajoene. Aster-
isks indicate the formation of large
cytoplasmic lipid droplets and
degeneration of the hepatocytes.
Arrows show vacuolar degenera-
tion and necrosis of hepatocytes. V,
large vacuoles. Scale bar, 100 mm
(
2
low magnification, upper panel) or
5 mm (high magnification, lower
panel). (C) Real-time PCR assays for
TNFa mRNA. The mRNA level
of GAPDH was used as a refer-
ence for data normalization. (D)
Immunoblottings for iNOS and
COX-2. iNOS and COX-2 were im-
munoblotted in the liver homoge-
nates of ND-fed or HFD-fed mice
that had been treated with vehicle
or 30 mg=kg=day ajoene. The sta-
tistical significance of differences
between each treatment group and
ND (*p < 0.05, **p < 0.01) or HFD
#
alone ( p < 0.05) was determined.
(
For interpretation of the references
to color in this figure legend, the
reader is referred to the web version
of this article at www.liebertonline
.com=ars).
6
B). The increase in SREBP-1c protein level by T090 was no- activity by ajoene contributes to the prevention of down-
tably antagonized by ajoene treatment. Consistently, treat- stream lipogenic gene induction in hepatocytes.
ment of ajoene decreased the T090-induced SREBP-1c mRNA
level (Fig. 6C). Moreover, the activities of other LXR agonists, Activation of AMPK that leads to S6K1 inhibition
2
2(R)-hydroxycholesterol as an endogenous ligand and
In view of the inhibitory role of AMPK activation in LXRa
activity and the resultant antisteatotic effects (18), the effect of
ajoene on AMPK activity was determined. The activity of
AMPK was assessed by monitoring the phosphorylation of
AMPKa and its best-characterized downstream kinase, ACC
GW3965 as an LXR-specific agonist, to induce SREBP-1c ex-
pression were also blocked by ajoene treatment (Fig. 6D).
Repression of SREBP-1c target gene induction
SREBP-1c induces the transcription of genes encoding for (30). In HepG2 cells, the levels of phospho-AMPKa and
lipogenic enzymes. Next, we examined whether the inhibition phospho-ACC were both notably increased by ajoene treat-
of SREBP-1c expression by ajoene causes the reduction of ment in a time-dependent manner (Fig. 8A, left).These in-
SREBP-1c activity to transactivate the target lipogenic genes. creases began from 10 min after ajoene treatment and were
Ajoene treatment attenuated the ability of the LXRa agonist to maintained at least up to 3 h. Treatment with ajoene enhanced
induce the activity of FAS reporter gene containing the sterol the phosphorylations of AMPKa and ACC in a concentration-
response element (SRE) in the ꢁ150 bp FAS promoter region dependent manner (Fig. 8A, right). Because AMPK can inhibit
(
Fig. 7A).Consistently, the T090-induced expression of lipo- the action of S6K1 that activates LXRa activity (18), the effect
genic genes, including FAS, ACC, and SCD-1, was signifi- of ajoene on S6K1 activity was additionally examined. As
cantly repressed by simultaneous treatment of ajoene (Fig. expected, ajoene treatment caused commensurate decreases
7B). These data show that the inhibition of LXRa-SREBP-1c in the phosphorylation of S6, a substrate of S6K1 (Fig. 8A).

194
HAN ET AL.
FIG. 4. Liver protection by ajoene from HFD-induced oxidative stress. (A) Thiobarbituric acid reactive substances
TBARS) assay. The amounts of TBARS were measured in liver homogenates. (B) Immunofluorescence for nitrotyrosine (red)
(
in the liver of ND-fed mice (ND) or HFD-fed mice (HFD) treated with vehicle or 10 mg=kg=day (Low) or 30 mg=kg=day (High)
0
ajoene. Arrows indicate the strong intensity of nitrotyrosine. Tissue sections were mounted with Vectashield containing 4 -6-
diamidino-2-phenylindole (DAPI) (blue) for nuclear staining. Pictures are representative results from at least eight experi-
ments. Scale bar, 50 mm. Ajo, ajoene; NT, nitrotyrosine. (C) Hepatic GSH content. The content of GSH was measured in liver
homogenates. (D) Hepatic GPx and catalase activities. The enzyme activities were measured in liver homogenates. Statistical
#
##
significance of differences between each treatment group and ND (*p < 0.05, **p < 0.01) or HFD alone ( p < 0.05, p < 0.01)
was determined. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of
this article at www.liebertonline.com=ars).
These time-dependent decreases reciprocally correlated with (Fig. 8E). It also inhibited the increased phosphorylation of S6.
ACC phosphorylation that depends on AMPK activation. S6 Our results support the roles of AMPK activation and S6K1
phosphorylation then returned towards the basal level at 3 h inhibition by ajoene in the antisteatotic effect both in vitro and
and 6 h. DAS and DADS, which can be made from allicin, in vivo. Collectively, ajoene has the capability to inhibit he-
have anticancer and cholesterol-lowering properties (19). We patic steatosis and protect the liver from oxidative injury,
comparatively evaluated the effects of the organosulfur which may result from the inhibition of LXRa-mediated li-
compounds on AMPK activation. Treatment of cells with ei- pogenesis via AMPK activation that leads to S6K1 inhibition.
ther 30 mM DAS or DADS only slightly increased ACC
phosphorylation (Fig. 8B), indicating that these agents are less
LKB1-dependent AMPK activation by ajoene
efficacious than ajoene.
To clarify whether repression of LXRa-SREBP-1c induction
AMPK may be activated by LKB1 or calcium=calmodulin-
by ajoene results from AMPK activation and S6K1 inhibition, dependent kinase kinase (CaMKK) through phosphorylation
the effects of DN-AMPKa or constitutively active (CA)-S6K1 (30). LKB1 phosphorylation was enhanced by ajoene treat-
transfection on SREBP-1c expression were examined. In ment in a time-dependent manner (Fig. 9A),which paralleled
HepG2 cells, either DN-AMPKa or CA-S6K1 antagonized the AMPK activation. Consistently, ajoene did not increase
ability of ajoene to inhibit T090-induced SREBP-1c expression phosphorylations of ACC and AMPKa in HeLa cells deficient
(Fig. 8C). Consistently, ajoene treatment suppressed T090- in LKB1 (Fig. 9B), confirming the role of LKB1 in the activation
induced LXRa binding to the LXRE region of the SREBP-1c of AMPK by ajoene. Similar results were obtained by siRNA
promoter, and which was reversed by AMPK inhibition (Fig. knockdown of LKB1 in HepG2 cells (Fig. 9C). AMPK activa-
8D). Moreover, ajoene prevented the repression in hepatic tion by ajoene was unchanged by treatment with STO-609, a
ACC and AMPK phosphorylations elicited by HFD-feeding CaMKKb inhibitor (Fig. 9D), suggesting that CaMKKb is not

INHIBITION OF HEPATIC STEATOSIS BY AJOENE
195
FIG. 5. Repression of hepatic lipogenic gene expression. (A) Real-time PCR assays for LXRa and SREBP-1c. (B) Real-time
PCR assays for FAS, ACC, and SCD-1. The mRNA level of GAPDH was used as a reference for data normalization. (C)
Plasma TG and total cholesterol contents. (D) Plasma fasting glucose level. Animals were treated as described in the legend to
#
Figure 2A. Statistical significance of differences between each treatment group and ND (**p < 0.01) or HFD alone ( p < 0.05,
#
#
p < 0.01) was determined. Ajo, ajoene.
involved in the process. Silent information regulator T1 tion, ajoene did not change cellular ATP level. These results
SIRT1) may affect AMPK activity. In an additional experi- demonstrate that AMPK activation by ajoene may depend on
(
ment, ajoene treatment (30 min) increased the ratio of the activation of LKB1 and SIRT1.
þ
NAD =NADH in HepG2 cells (Fig. 9E). Moreover, ajoene-
induced ACC and AMPKa phosphorylations were both
abolished by treatment of 10 mM nicotinamide or 10 mM sir-
Discussion
tinol (i.e., SIRT1 inhibitors), suggesting that SIRT1 may be
Hepatic steatosis is emerging as the most common patho-
required for the LKB1-dependent AMPK activation. In addi- logic state of liver with the growing incidence of overweight

196
HAN ET AL.
FIG. 6. The effect of ajoene on LXRa-SREBP-1c activation in hepatocytes. (A) Ajoene inhibition of LXRa activation.
HepG2 cells were treated with vehicle or 10 mM T0901317 (T090) in combination with ajoene for 12 h. The LXRa mRNA level
was analyzed using real-time RT-PCR assay (left), with that of GAPDH used as a normalizing reference. The relative LXRE
luciferase activity (right) was monitored in the lysates of cells exposed to different treatment combinations (T090 for 12 h with
or without ajoene pretreatment for 1 h). (B) The expression of SREBP-1c protein. Immunoblots were performed on the lysates
of HepG2 cells or primary rat hepatocytes that had been treated with ajoene (10 mM or 30 mM) for 1 h followed by subsequent
incubation with T090 for 12 h. Immunoblots are representative results from at least 3 experiments. (C) Real-time RT-PCR
assays for SREBP-1c in HepG2 cells. (D) The effects of ajoene on SREBP-1c induction by 22(R)-hydroxycholesterol [22(R)-HC,
1
0 mM] or GW3965 (10 mM) treatment for 12 h. Data represent the mean ꢂ S.E. of three replicates. Statistical significance of
#
##
differences between each treatment group and the control (**p < 0.01) or T090 without ajoene treatment ( p < 0.05, p < 0.01)
was determined. Ajo, ajoene.

INHIBITION OF HEPATIC STEATOSIS BY AJOENE
197
FIG. 7. Inhibition by ajoene of SREBP-1c-dependent gene induction. (A) Ajoene inhibition of T090-induced, SRE-
mediated FAS gene transactivation. FAS luciferase assays were performed on the lysates of HepG2 cells that had been treated
as described in the legend to Figure 6A. (B) Real-time PCR assays. HepG2 cells were treated with vehicle or T090 in
combination with ajoene for 12 h. The transcripts of lipogenic genes were analyzed as described in the legend to Figure 6A.
The GAPDH mRNA level was used as a normalizing reference. Data represent the mean ꢂ S.E. of three replicates. Statistical
#
significance of differences between each treatment group and the control (**p < 0.01) or T090 without ajoene ( p < 0.05,
#
#
p < 0.01) was determined. Ajo, ajoene.
and obesity (25). Steatosis in the development of NAFLD is a observation proved obvious fat accumulation in the liver after
strong risk factor for insulin resistance mostly observed in feeding of mice with 60% HFD for 8 weeks, which was evi-
obese people. Overproduction and deposition of fat in hepa- denced by increases in Oil Red O staining and hepatic TG
tocytes is the primary event that stimulates progression of content. These phenotypic alterations were clearly antago-
metabolic liver diseases (3). Moreover, it may progress to nized by ajoene treatment (10 or 30 mg=kg=day) for the final 4
advanced liver diseases with associated morbidity and mor- weeks of the HFD feeding. Moreover, ajoene significantly
tality: steatohepatitis and more severely to fibrosis, cirrhosis, reduced the increased plasma TG and cholesterol levels,
and hepatocellular carcinoma (35). The present study was which is in line with the report that it inhibits cholesterol
designed to determine the effects of ajoene on lipogenesis and synthesis in hepatocytes (14, 43).
hepatic steatosis using both an in vivo model involving mice
fed an HFD and an in vitro model using a treatment of LXRa was diminished by treatment of 30 mg=kg=day ajoene. Al-
ligand on hepatocytes. though a low dose (10 mg=kg=day) of ajoene treatment non-
In addition, increased body weight gain in HFD-fed mice
Since HFD feeding accompanies obesity and insulin resis- significantly affected body weight gain, there was a trend to
tance in animals, inducing fatty liver, the pathophysiologic decrease it. So, successful inhibition of hepatic lipid accumu-
features in animals are analogous to clinical metabolic dis- lation by 10 mg=kg=day of ajoene treatment may be associ-
eases. Although rodents fed a methionine- and choline- ated in part with anti-obesity action. Body weight gain is
deficient diet develop hepatic steatosis with inflammation, affected by the balance between energy supplement and
this diet feeding causes loss of body weight gain (e.g., muscle consumption (26). Our data showing no change in the food
and fat) and does not develop insulin resistance (35). The intake after ajoene treatment suggest the possibility that
period of HFD feeding and the amount of fat in the diet affect ajoene has an effect on energy expenditure process. Interest-
the metabolic phenotypes and liver histology. Usually, 40%– ingly, ajoene inhibits differentiation of preadipocytes and also
60% HFD feeding (% calories from fat) are used; body weight induces adipocyte apoptosis (38, 53), which may account for
and plasma glucose level were both higher in 58% HFD-fed its anti-obesity effect. Ajoene may also change dietary lipid
mice already after the first week (51), and liver fat levels in- absorption, as suggested by the finding that it inhibited hu-
creased within a week by this diet feeding (41). Indeed, our man gastric lipase in vitro (13). Moreover, ajoene treatment

1
98
HAN ET AL.
FIG. 8. The ability of ajoene
to activate AMPK that leads
to S6K1 inhibition. (A)
AMPK activation and inhibi-
tion of S6 phosphoryla-
tion. Immunoblots were
performed on the lysates of
HepG2 cells that had been
treated with ajoene (30 mM)
for the indicated time period
or different concentration of
ajoene (3–30 mM) for 30 min.
(
B) The effects of DAS and
DADS on ACC phosphory-
lation. Immunoblots were
performed on the lysates of
HepG2 cells that had been
treated with either DAS or
DADS (30 mM each) for the
indicated time period. (C)
The roles of AMPK and S6K1
in the ability of ajoene to in-
hibit SREBP-1c induction by
T090. Immunoblots were
performed on the lysates of
HepG2 cells that had been
treated with T090 or T090þ
ajoene for 12 h following
DN-AMPK
transfection. (D) ChIP assays.
DNA–protein complexes
or
CA-S6K1
were prepared from cells that
had been treated with T090
or T090þajoene for 6 h
following DN-AMPK trans-
fection, and were immuno-
precipitated with anti-LXRa
antibody or preimmune-IgG.
The samples were subjected
to PCR-amplification using
DNA primers flanking the
LXRE of SREBP-1c gene pro-
moter. (E) Immunoblottings
for phosphorylated ACC,
AMPK, and S6. Proteins of
interest were immunoblotted
in the liver homogenates of
ND-fed or HFD-fed mice that
had been treated with vehicle
or ajoene (30 mg=kg=day).
Data represent the mean ꢂ
S.E. of three replicates. Sta-
tistical significance of differ-
ences between each treatment
group and the control
(
*p < 0.05, **p < 0.01) or T090
#
#
without ajoene ( p < 0.01)
was determined. Ajo, ajoene.
improved hyperglycemia and hyperinsulinemia induced by reactive oxygen species (ROS) in hepatocytes. Thus, oxidative
HFD feeding. Taken together, our results demonstrate that stress is one of the key stimuli that mediate lipotoxic events
ajoene has the ability to inhibit hepatic steatosis, decrease and evoke pathologic development (34, 35); the radical stress
body weight gain, and improve glucose metabolism.
triggers further production of lipid peroxides that make ad-
Lipid-induced toxicity caused by fat overload may cause ducts to macromolecules, resulting in chronic damage to or-
steatohepatitis (i.e., liver injury), which is a crucial step for ganelles such as mitochondria (35). The degeneration and
progression to end-stage liver disease (35). Excessive oxida- apoptosis of parenchymal cells may lead to collapse of organ
tion of accumulated fatty acids generates large amounts of function. Moreover, oxidative stress stimulates the release of

INHIBITION OF HEPATIC STEATOSIS BY AJOENE
199
FIG.
AMPK activation by ajoene.
A) LKB1 phosphorylation.
Immunoblots were per-
9. LKB1-dependent
(
formed on the lysates of
HepG2 cells that had been
treated with 30 mM ajoene for
the indicated times. (B)
AMPK activation. Hela cells
deficient in LKB1 were trea-
ted with vehicle or ajoene for
1
h and subjected to immu-
noblot assay. (C) Effect of
LKB1 siRNA knockdown.
HepG2 cells were treated
with ajoene for 1 h after siR-
NA transfection. (D) Effect of
CaMKKb inhibitor on the ac-
tivation of AMPK. HepG2
cells were treated with
1
mg=ml STO-609 for 30 min,
and continuously incubated
with ajoene for 1 h. (E) Cel-
þ
lular NAD =NADH ratio.
þ
The contents of NAD and
NADH were measured in
cells treated with 30 mM
ajoene for 30 min. (F) Effects
of SIRT1 inhibitors on the ac-
tivation of AMPK by ajoene.
HepG2 cells were treated
with 10 mM nicotinamide
(
3
NAM) or 10 mM sirtinol for
0 min, and continuously in-
cubated with ajoene for 1 h.
G) A schematic diagram il-
lustrating the proposed
(
mechanism by which ajoene
inhibits LXRa-mediated he-
patic steatosis. Ajo, ajoene.
inflammatory mediators. Imbalance of redox state and ticular, ajoene as one of organosulfur by-products derived
chronic inflammatory condition alter intracellular signaling, from garlic has beneficial effects and chemical stability. Along
and thus impair energy metabolism (22, 50). Therefore, oxi- with numerous studies of garlic on atherosclerosis and cancer,
dative stress is critical for the development of NAFLD. In this the hepatoprotective effect of garlic against toxicants has been
study, HFD feeding indeed promoted oxidative stress, as also recognized (9, 10, 31). Our results shown here demon-
shown by increases in TBARS and nitrotyrosinylation (i.e., strate for the first time that ajoene treats fatty liver and fat-
ROS and RNS) with a decrease in cellular GSH content that induced oxidative stress, and which results mainly from the
was accompanied by hepatocytes injury (e.g., liver histopa- inhibition of lipogenesis.
thology, inflammatory gene induction, and plasma ALT ac-
A line of evidence indicates that abnormal hepatic fat ac-
tivity). Intriguingly, ajoene treatment reversed these cumulation in NAFLD patients results largely from de novo
pathologic changes: reduction of free radical stress and in- lipogenesis (8, 45). LXRa is highly expressed in liver, acting as
flammatory response to damage. Our findings showed that a lipid sensing nuclear receptor to induce fat synthesis mainly
ajoene attenuates steatohepatitis induced by fat accumula- through transcription of SREBP-1c, a master regulator of li-
tion, and that the beneficial effect may be associated with its pogenic genes (39). The lipogenic enzymes including FAS,
inhibitory action on oxidative stress.
ACC, and SCD-1 (i.e., targets of SREBP-1c) are involved in
Garlic has been used as a remedy for disease as well as a multiple steps of TG anabolic processes (3). The importance of
food ingredient. The favorable actions of garlic are elicited LXRa in steatosis was proven by the findings that adminis-
mostly by diverse organosulfur components (15, 19). In par- tration of LXR activator to mice induces hepatic lipogenesis

2
00
HAN ET AL.
and TG accumulation, whereas a deficiency of LXR abolished upstream kinase of AMPK activation elicited by ajoene.
this effect (42). Consistently, HFD feeding increased hepatic Emerging evidence supports that SIRT1, an NAD-dependent
and plasma TG levels in mice, whereas no increase was found deacetylase, collaborates with AMPK to regulate energy me-
in LXR-null mice (21). In the present study, ajoene effica- tabolism (16). Our results showing an increase in
þ
ciously antagonized LXRa and SREBP-1c gene expressions NAD =NADH by ajoene and inhibition of ajoene-induced
induced by HFD feeding in the liver tissue of mice, as further AMPK activation by SIRT1 inhibitors support the concept that
supported by the repression of lipogenic genes, FAS, ACC, SIRT1 is necessary for the LKB1-dependent activation of
and SCD-1.
AMPK by ajoene. In addition, AMPK induces small hetero-
Consistent with the results of an in vivo experimental dimer partner (SHP) by activating upstream stimulatory fac-
model, ajoene inhibited LXRa agonist-induced SREBP-1c ac- tor-1 (5); SHP then inhibits the SREBP-1c lipogenic pathway
tivation and lipogenic gene induction in hepatocytes, as through LXRa inhibition (49). In a supplemental experiment,
shown by the results of immunoblot, real-time PCR, and re- ajoene increased SHP mRNA level (*1.6-fold), supporting
porter gene assays. The activity of LXRa is post-translationally the possible role of SHP in the action of ajoene. Moreover,
controlled by phosphorylation. Recently, Hwahng et al. AMPK has the ability to protect mitochondria from oxidative
showed that AMPK and S6K1 oppositely regulate LXRa ac- stress in association with glycogen synthase kinase-3b phos-
tivity: the inhibitory and activating phosphorylations of LXRa phorylation (6).
are mediated by AMPK and S6K1, respectively (18). In the
Pharmacological interventions may achieve the therapeutic
present study, the inhibitory effect of ajoene on LXRa induc- effects to prevent and=or to treat steatosis and steatohepatitis.
tion of SREBP-1c was reversed by either AMPK inhibition or Although insulin sensitizers are favorably used in clinical
S6K1 activation, indicating that the antilipogenic effect may situations, some drugs, including thiazolidinediones, stimu-
stem from LXRa inhibition as a consequence of AMPK acti- late weight gain, which is an unwanted effect in NAFLD pa-
vation that leads to S6K1 inhibition. Other kinases that tients (11). Various antioxidants including radical scavengers
modulate the activity of LXRa include protein kinase A and are used to treat or delay the progression of steatosis. In our
casein kinase-2 (47, 52). The effects of ajoene on these remain study, we found no direct radical scavenging effect of ajoene,
to be explored.
as assessed by an experiment using 1,1-diphenyl-2-picrylhy-
In addition to the antilipogenic effect of AMPK, it also drazyl. So, it is likely that the antioxidant effect of ajoene re-
regulates cellular lipid metabolism in large part through sults from its ability to activate AMPK. In addition, we
stimulation of fatty acid oxidation (48). We found that ajoene recently reported that ajoene induces antioxidant genes via
treatment increased the level of hepatic CPT-1, which is a Nrf2 activation (24). Moreover, ajoene inhibits the induction
downstream effecter of peroxisome proliferator-activated re- of inflammatory genes in macrophages (7), and has anti-
ceptor (PPAR)-a [HFD alone, 1.00 ꢂ 0.08; HFDþAjo platelet and anti-atherosclerosis effects (1). All of these effects
1
1
0 mg=kg=day, 1.45 ꢂ 0.16*; and HFDþAjo 30 mg=kg=day, add value to the potential of ajoene for the treatment of
.83 ꢂ 0.16** (statistical significance of difference from HFD, steatosis and steatohepatitis. Our finding that ajoene inhibits
*
p < 0.05, **p < 0.01)]. AMPK has also been shown to be in- fat-induced radical stress further strengthens its potential
volved in the control of mitochondrial biogenesis and adap- application for steatohepatitis. Easy synthesis and chemical
tive mitochondrial function (40). In mice fed on an HFD, stability are also merits of this compound as a drug candidate.
ajoene increased the level of hepatic PPARg coactivator-1a
(
PGC-1a) mRNA [ND, 1.00 ꢂ 0.16; HFD alone, 0.45 ꢂ 0.04**; Acknowledgments
#
and HFDþAjo 10 mg=kg=day, 0.64 ꢂ 0.06 (statistical signifi-
This work was supported by the National Research Foun-
dation of Korea grant funded by the Korea government
cance of difference from ND, **p < 0.01 or HFD alone,
#
p < 0.05)], suggesting that the activated AMPK upregulates
(
MEST)(No. 2010-0001706), Korea. CYH would like to thank
PGC-1a and promotes mitochondrial biogenesis. Collectively,
our results indicate that ajoene may have an additional effect
to increase fatty acid oxidation in the liver.
AMPK activation exerts beneficial effects such as anti-
steatosis and antidiabetes, which may also be caused by its
inhibition of S6K1 (2, 18). In the present study, phosphoryla-
the Seoul Science Fellowship program.
Author Disclosure Statement
No competing financial interests exist.
tions of AMPK and ACC were both increased by ajoene References
treatment in hepatocytes with a compatible decrease in S6K1
1
. Apitz–Castro R, Badimon JJ, and Badimon L. Effect of
activity; moreover, ajoene was more efficacious than other
organosulfurs (i.e., DAS and DADS). Therefore this study
identified ajoene as a component that regulates AMPK-S6K1
pathway and thereby represses LXRa activity. Other AMPK
activators (e.g., dithiolethione and sauchinone) have an anti-
steatotic effect as well, which is mediated by the inhibition of
LXRa (18, 28). LXRa may have a favorable effect on cholesterol
and glucose metabolism (32). So it cannot be excluded that the
beneficial effect of ajoene on plasma cholesterol, glucose, and
insulin results from the AMPK regulation of other cascades
apart from LXRa inhibition.
ajoene, the major antiplatelet compound from garlic, on
platelet thrombus formation. Thromb Res 68: 145–155, 1992.
. Bae EJ, Yang YM, Kim JW, and Kim SG. Identification of a
novel class of dithiolethiones that prevent hepatic insulin
resistance via the adenosine monophosphate-activated pro-
tein kinase-p70 ribosomal S6 kinase-1 pathway. Hepatology
2
4
6: 730–739, 2007.
3
. Browning JD and Horton JD. Molecular mediators of hepatic
steatosis and liver injury. J Clin Invest 114: 147–152, 2004.
4. Buege JA and Aust SD. Microsomal lipid peroxidation.
Methods Enzymol 52: 302–310, 1978.
In the present study, ajoene activated AMPK through
LKB1, but not CaMKK, implying that LKB1 may serve as an
5. Chanda D, Li T, Song KH, Kim YH, Sim J, Lee CH, Chiang
JY, and Choi HS. Hepatocyte growth factor family nega-

INHIBITION OF HEPATIC STEATOSIS BY AJOENE
201
tively regulates hepatic gluconeogenesis via induction of 23. Kang KW, Kim YG, Cho MK, Bae SK, Kim CW, Lee MG, and
orphan nuclear receptor small heterodimer partner in pri-
mary hepatocytes. J Biol Chem 284: 28510–28521, 2009.
. Choi SH, Kim YW, and Kim SG. AMPK-mediated GSK3beta
Kim SG. Oltipraz regenerates cirrhotic liver through
CCAAT=enhancer binding protein-mediated stellate cell in-
activation. FASEB J 16: 1988–1990, 2002.
6
inhibition by isoliquiritigenin contributes to protecting mi- 24. Kay HY, Yang JW, Kim TH, Lee DY, Kang B, Ryu JH, Jeon R,
tochondria against iron-catalyzed oxidative stress. Biochem
Pharmacol 79: 1352–1362, 2010.
and Kim SG. Ajoene, a stable garlic by-product, has an an-
tioxidant effect through Nrf2-mediated glutamate-cysteine
ligase induction in HepG2 cells and primary hepatocytes. J
Nutr 140: 1211–1219, 2010.
7
. Dirsch VM, Kiemer AK, Wagner H, and Vollmar AM. Effect
of allicin and ajoene, two compounds of garlic, on inducible
nitric oxide synthase. Atherosclerosis 139: 333–339, 1998.
. Donnelly KL, Smith CI, Schwarzenberg SJ, Jessurun J, Boldt
MD, and Parks EJ. Sources of fatty acids stored in liver and
25. Kotronen A and Yki–J a¨ rvinen H. Fatty liver: A novel com-
ponent of the metabolic syndrome. Arterioscler Thromb Vasc
Biol 28: 27-38, 2008.
8
secreted via lipoproteins in patients with nonalcoholic fatty 26. Kim MS, Park JY, Namkoong C, Jang PG, Ryu JW, Song HS,
liver disease. J Clin Invest 115: 1343–1351, 2005.
Yun JY, Namgoong IS, Ha J, Park IS, Lee IK, Viollet B, Youn
JH, Lee HK, and Lee KU. Anti-obesity effects of alpha-lipoic
acid mediated by suppression of hypothalamic AMP-
activated protein kinase. Nat Med 10: 727–733, 2004.
27. Kim YW, Kang HE, Lee MG, Hwang SJ, Kim SC, Lee CH,
and Kim SG. Liquiritigenin, a flavonoid aglycone from lic-
orice, has a choleretic effect and the ability to induce hepatic
transporters and phase-II enzymes. Am J Physiol Gastrointest
Liver Physiol 296: G372–381, 2009.
9
. El–Beshbishy HA. Aqueous garlic extract attenuates hep-
atitis and oxidative stress induced by galactosamine=
lipoploysaccharide in rats. Phytother Res 22: 1372–1379,
2
008.
1
0. Flora SJ, Mehta A, and Gupta R. Prevention of arsenic-in-
duced hepatic apoptosis by concomitant administration of
garlic extracts in mice. Chem Biol Interact 177: 227–233, 2009.
1. Fonseca V. Effect of thiazolidinediones on body weight in
1
1
patients with diabetes mellitus. Am J Med 115: 42S–48S, 2003. 28. Kim YW, Kim YM, Yang YM, Kim TH, Hwang SJ, Lee JR,
2. Fujisawa H, Suma K, Origuchi K, Kumagai H, Seki T, and
Ariga T. Biological and chemical stability of garlic-derived
allicin. J Agric Food Chem 56: 4229–4235, 2008.
Kim SC, and Kim SG. Inhibition of SREBP-1c-mediated he-
patic steatosis and oxidative stress by sauchinone, an
AMPK-activating lignan in Saururus chinensis. Free Radic Biol
Med 48: 567–578, 2010.
1
1
1
1
3. Gargouri Y, Moreau H, Jain MK, de Haas GH, and Verger R.
Ajoene prevents fat digestion by human gastric lipase in 29. Laffitte BA, Joseph SB, Walczak R, Pei L, Wilpitz DC, Collins
vitro. Biochim Biophys Acta 1006: 137–139, 1989.
JL, and Tontonoz P. Autoregulation of the human liver X
4. Gebhardt R, Beck H, and Wagner KG. Inhibition of choles-
receptor alpha promoter. Mol Cell Biol 21: 7558–7568, 2001.
terol biosynthesis by allicin and ajoene in rat hepatocytes 30. Lage R, Di e´ guez C, Vidal–Puig A, and L o´ pez M. AMPK: A
and HepG2 cells. Biochim Biophys Acta 1213: 57–62, 1994.
metabolic gauge regulating whole-body energy homeostasis.
5. Hassan HT. Ajoene (natural garlic compound): A new anti-
Trends Mol Med 14: 539–549, 2008.
leukaemia agent for AML therapy. Leuk Res 28: 667–671, 31. Lee SC, Hwang SY, Kim SW, and Kim SK. Ethanol extract of
2
004.
Allium sativum attenuates testicular and liver toxicity in-
duced by 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. J Med
Food 12: 93–99, 2009.
6. Hou X, Xu S, Maitland-Toolan KA, Sato K, Jiang B, Ido Y,
Lan F, Walsh K, Wierzbicki M, Verbeuren TJ, Cohen RA,
and Zang M. SIRT1 regulates hepatocyte lipid metabolism 32. Lehrke M, Lebherz C, Millington SC, Guan HP, Millar J,
through activating AMP-activated protein kinase. J Biol
Chem 283: 20015–20026, 2008.
7. Hunter R, Kaschula CH, Parker IM, Caira MR, Richards P,
Rader DJ, Wilson JM, and Lazar MA. Diet-dependent car-
diovascular lipid metabolism controlled by hepatic LXRal-
pha. Cell Metab 1: 297–308, 2005.
1
1
Travis S, Taute F, and Qwebani T. Substituted ajoenes as 33. Liang G, Yang J, Horton JD, Hammer RE, Goldstein JL, and
novel anti-cancer agents. Bioorg Med Chem Lett 18: 5277–
279, 2008.
Brown MS. Diminished hepatic response to fasting=refeed-
ing and liver X receptor agonists in mice with selective de-
ficiency of sterol regulatory element-binding protein-1c. J
Biol Chem 277: 9520–9528, 2002.
5
8. Hwahng SH, Ki SH, Bae EJ, Kim HE, and Kim SG. Role of
adenosine monophosphate-activated protein kinase-p70 ri-
bosomal S6 kinase-1 pathway in repression of liver X re- 34. Mantena SK, King AL, Andringa KK, Eccleston HB, and
ceptor-alpha-dependent lipogenic gene induction and
hepatic steatosis by a novel class of dithiolethiones. Hepa-
tology 49: 1913–1925, 2009.
Bailey SM. Mitochondrial dysfunction and oxidative stress
in the pathogenesis of alcohol- and obesity-induced fatty
liver diseases. Free Radic Biol Med 44: 1259–1272, 2008.
1
2
2
9. Iciek M, Kwiecie n´ I, and Włodek L. Biological properties of 35. Marra F, Gastaldelli A, Svegliati Baroni G, Tell G, and Tir-
garlic and garlic-derived organosulfur compounds. Environ
Mol Mutagen 50: 247–265, 2009.
0. Inoki K, Zhu T, and Guan KL. TSC2 mediates cellular energy
ibelli C. Molecular basis and mechanisms of progression of
non-alcoholic steatohepatitis. Trends Mol Med 14: 72–81,
2008.
response to control cell growth and survival. Cell 115: 577– 36. Meister A. Glutathione metabolism and its selective modi-
90, 2003. fication. J Biol Chem 263: 17205–17208, 1988.
1. Kalaany NY, Gauthier KC, Zavacki AM, Mammen PP, Ki- 37. Naganawa R, Iwata N, Ishikawa K, Fukuda H, Fujino T, and
5
tazume T, Peterson JA, Horton JD, Garry DJ, Bianco AC, and
Mangelsdorf DJ. LXRs regulate the balance between fat
storage and oxidation. Cell Metab 1: 231–244, 2005.
Suzuki A. Inhibition of microbial growth by ajoene, a sulfur-
containing compound derived from garlic. Appl Environ
Microbiol 62: 4238–4242, 1996.
2
2. Kaneto H, Katakami N, Kawamori D, Miyatsuka T, Saka- 38. Nelson–Dooley C, Della–Fera MA, Hamrick M, and Baile
moto K, Matsuoka TA, Matsuhisa M, and Yamasaki Y. In-
volvement of oxidative stress in the pathogenesis of
diabetes. Antioxid Redox Signal 9: 355–366, 2007.
CA. Novel treatments for obesity and osteoporosis: Target-
ing apoptotic pathways in adipocytes. Curr Med Chem 12:
2215–2225, 2005.

2
02
HAN ET AL.
39. Repa JJ, Liang G, Ou J, Bashmakov Y, Lobaccaro JM, Shi-
phosphorylation of liver X receptor in the liver. J Biol Chem
momura I, Shan B, Brown MS, Goldstein JL, and Man-
282: 11687–11695, 2007.
gelsdorf DJ. Regulation of mouse sterol regulatory element- 53. Yang JY, Della–Fera MA, Hausman DB, and Baile CA. En-
binding protein-1c gene (SREBP-1c) by oxysterol receptors,
LXRalpha and LXRbeta. Genes Dev 14: 2819–2830, 2000.
0. Reznick RM and Shulman GI. The role of AMP-activated
protein kinase in mitochondrial biogenesis. J Physiol 574: 33–
hancement of ajoene-induced apoptosis by conjugated lino-
leic acid in 3T3-L1 adipocytes. Apoptosis 12: 1117–1128, 2007.
4
4
Address correspondence to:
Sang Geon Kim, Ph.D.
College of Pharmacy
3
9, 2006.
1. Samuel VT, Liu ZX, Qu X, Elder BD, Bilz S, Befroy D, Ro-
manelli AJ, and Shulman GI. Mechanism of hepatic insulin
resistance in non-alcoholic fatty liver disease. J Biol Chem
and Research Institute of Pharmaceutical Sciences
Seoul National University
Sillim-dong, Kwanak-gu
Seoul 151-742
2
79: 32345–32353, 2004.
4
4
4
2. Schultz JR, Tu H, Luk A, Repa JJ, Medina JC, Li L,
Schwendner S, Wang S, Thoolen M, Mangelsdorf DJ, Lustig
KD, and Shan B. Role of LXRs in control of lipogenesis.
Genes Dev 14: 2831–2838, 2000.
3. Sendl A, Schliack M, L o¨ ser R, Stanislaus F, and Wagner H.
Inhibition of cholesterol synthesis in vitro by extracts and
isolated compounds prepared from garlic and wild garlic.
Atherosclerosis 94: 79–85, 1992.
4. Stadler K, Bonini MG, Dallas S, Jiang J, Radi R, Mason RP,
and Kadiiska MB. Involvement of inducible nitric oxide
synthase in hydroxyl radical-mediated lipid peroxidation
in streptozotocin-induced diabetes. Free Radic Biol Med 45:
Korea
E-mail: sgk@snu.ac.kr
Date of first submission to ARS Central, March 17, 2010; date
of final revised submission, May 27, 2010; date of acceptance,
June 19, 2010.
Abbreviations Used
ACC ¼ acetyl-CoA carboxylase
ALT ¼ alanine aminotransferase
AMPK ¼ AMP-activated protein kinase
CA ¼ constitutively active
8
66–874, 2008.
4
4
5. Tamura S and Shimomura I. Contribution of adipose tissue
and de novo lipogenesis to nonalcoholic fatty liver disease. J
Clin Invest 115: 1139–1142, 2005.
6. Taylor P, Noriega R, Farah C, Abad MJ, Arsenak M, and
Apitz R. Ajoene inhibits both primary tumor growth and
metastasis of B16=BL6 melanoma cells in C57BL=6 mice.
Cancer Lett 239: 298–304, 2006.
7. Torra IP, Ismaili N, Feig JE, Xu CF, Cavasotto C, Pancratov
R, Rogatsky I, Neubert TA, Fisher EA, and Garabedian MJ.
Phosphorylation of liver X receptor alpha selectively regu-
lates target gene expression in macrophages. Mol Cell Biol 28:
CaMKK ¼ calcium=calmodulin-dependent kinase kinase
COX-2 ¼ cyclooxygenase-2
CPT-1 ¼ carnitine palmitoyl transferase-1
DADS ¼ diallyl disulfide
DAS ¼ diallyl sulfide
4
DN ¼ dominant negative
FAS ¼ fatty acid synthase
GPx ¼ glutathione peroxidase
HFD ¼ high-fat diet
2
626–2636, 2008.
iNOS ¼ inducible nitric oxide synthase
LKB1 ¼ liver kinase B1
4
8. Viollet B, Mounier R, Leclerc J, Yazigi A, Foretz M, and
Andreelli F. Targeting AMP-activated protein kinase as a
novel therapeutic approach for the treatment of metabolic
disorders. Diabetes Metab 33: 395–402, 2007.
9. Watanabe M, Houten SM, Wang L, Moschetta A, Man-
gelsdorf DJ, Heyman RA, Moore DD, and Auwerx J. Bile
acids lower triglyceride levels via a pathway involving FXR,
SHP, and SREBP-1c. J Clin Invest 113: 1408–1418, 2004.
0. Wellen KE and Hotamisligil GS. Inflammation, stress, and
diabetes. J Clin Invest 115: 1111–1119, 2005.
1. Winzell MS and Ahr e´ n B. The high-fat diet-fed mouse: A
model for studying mechanisms and treatment of impaired
glucose tolerance and type 2 diabetes. Diabetes 53 Suppl 3:
S215–S219, 2004.
2. Yamamoto T, Shimano H, Inoue N, Nakagawa Y, Matsu-
zaka T, Takahashi A, Yahagi N, Sone H, Suzuki H,
Toyoshima H, and Yamada N. Protein kinase A suppresses
sterol regulatory element-binding protein-1C expression via
LXRa ¼ liver X receptor-a
LXRE ¼ LXR response element
m-CPBA ¼ meta-chloroperoxybenzoic acid
NAFLD ¼ nonalcoholic fatty liver disease
ND ¼ normal diet
4
PGC-1a ¼ PPARg coactivator-1a
PPAR ¼ peroxisome proliferator-activated receptor
RNS ¼ reactive nitrogen species
ROS ¼ reactive oxygen species
SCD ¼ stearoyl-CoA desaturase
SHP ¼ small heterodimer partner
SIRT1 ¼ silent information regulator T1
SREBP-1c ¼ sterol regulatory element binding protein-1c
T090 ¼ T0901317
5
5
5
TBARS ¼ thiobarbituric acid reactive substances
TG ¼ triglyceride