Anti-inflammatory mechanisms of isoflavone metabolites in lipopolysaccharide-stimulated microglial cells

Article abstract of DOI:10.1124/jpet.106.114322

The microglial activation plays an important role in neurodegenerative diseases by producing several proinflammatory cytokines and nitric oxide (NO). We found that three types of isoflavones and their metabolites that are transformed by the human intestin

Full text of DOI:10.1124/jpet.106.114322

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
0022-3565/07/3203-1237–1245$20.00
THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS
Copyright © 2007 by The American Society for Pharmacology and Experimental Therapeutics
JPET 320:1237–1245, 2007
Vol. 320, No. 3
114322/3185269
Printed in U.S.A.
Anti-Inflammatory Mechanisms of Isoflavone Metabolites in
Lipopolysaccharide-Stimulated Microglial Cells
Jin-Sun Park, Moon-Sook Woo, Dong-Hyun Kim, Jin-Won Hyun, Won-Ki Kim,
Jae-Chul Lee, and Hee-Sun Kim
Department of Neuroscience and Medical Research Institute, College of Medicine (J.-S.P., M.-S.W., H.-S.K.) and Division of
NanoScience (W.-K.K., J.-C.L.), Ewha Womans University, Seoul, Republic of Korea; Department of Microbial Chemistry,
College of Pharmacy, Kyung Hee University, Seoul, Republic of Korea (D.-H.K.); and Department of Biochemistry, College of
Medicine, Cheju National University, Jeju, Republic of Korea (J.-W.H.)
Received September 19, 2006; accepted December 22, 2006
ABSTRACT
The microglial activation plays an important role in neurode- Irisolidone significantly inhibited the DNA binding and transcrip-
generative diseases by producing several proinflammatory cy- tional activity of nuclear factor (NF)-␬B and activator protein-1.
tokines and nitric oxide (NO). We found that three types of Moreover, it repressed the LPS-induced extracellular signal-
isoflavones and their metabolites that are transformed by the regulated kinase (ERK) phosphorylation without affecting the
human intestinal microflora suppress lipopolysaccharide (LPS)- activity of c-Jun N-terminal kinase or p38 mitogen-activated
induced release of NO and tumor necrosis factor (TNF)-␣ in protein kinase. The level of NF-␬B inhibition by irisolidone cor-
primary cultured microglia and BV2 microglial cell lines. The related with the level of iNOS, TNF-␣, and interleukin (IL)-1␤
inhibitory effect of the isoflavone metabolites (aglycon form) suppression in LPS-stimulated microglia, whereas the level of
was more potent than that of isoflavones (glycoside form). The ERK inhibition correlated with the level of TNF-␣ and IL-1␤
RNase protection assay showed that the isoflavone metabo- repression. Overall, the repression of proinflammatory cyto-
lites regulated inducible nitric oxide synthase (iNOS) and the kines and iNOS gene expression in activated microglia by
cytokines at either the transcriptional or post-transcriptional isoflavones such as irisolidone might have therapeutic potential
level. A further molecular mechanism study was performed for for various neurodegenerative diseases including ischemic ce-
irisolidone, a metabolite of kakkalide, which had the most po- rebral disease.
tent anti-inflammatory effect among the six isoflavones tested.
Microglia are the primary immune cells of the brain that at least two functionally distinct states, namely a phagocytic
are activated in response to brain injury and release of
various neurotrophic factors supporting neuronal cell sur-
vival or neurotoxic factors, including nitric oxide (NO) and
proinflammatory cytokines such as TNF-␣ and IL-1␤
(Kreutzberg, 1996). There is accumulating evidence suggest-
ing that microglial activation is not just a single phenotype
but is also a complex array of responses (Town et al., 2005).
It was suggested that once activated, microglial cells exist in
phenotype (innate activation) or an antigen-presenting phe-
notype (adaptive activation), as determined by their stimu-
latory environment. The imbalance of these states can render
microglial activation either beneficial or detrimental to
neurons.
A number of studies have reported that neuronal cell death
causes microglial activation by releasing various signaling
molecules (Aldskogius et al., 1999; Kim et al., 2005), which
suggests that neuroinflammation is the result of an ongoing
disease process. On the other hand, several studies have
suggested that activated microglia play a role as a possible
cause of Alzheimer’s disease by secreting proinflammatory
cytokines such as TNF-␣ and IL-1␤, which have been shown
to promote neuronal injury at high levels (Tan et al., 1999,
This work was supported by the Health Planning Technology and Evalua-
tion Board (Grant A050451). J.S.P. was supported by the Brain Korea 21
project in 2006.
Article, publication date, and citation information can be found at
http://jpet.aspetjournals.org.
doi:10.1124/jpet.106.114322.
ABBREVIATIONS: NO, nitric oxide; TNF, tumor necrosis factor; IL, interleukin; LPS, bacterial lipopolysaccharide; NF, nuclear factor; ERK,
extracellular signal-regulated kinase; PD98059, 2Ј-amino-3Ј-methoxyflavone; U0126, 1,4-diamino-2,3-dicyano-1,4-bis(2-aminophenylthio)buta-
diene; PDTC, pyrrolidone dithiocarbamate; MG132, carbobenzoxy-L-Leucyl-L-leucyl-L-lucinal; MEM, modified Eagle’s medium; FCS, fetal calf
serum; MTT, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide; ELISA, enzyme-linked immunosorbent assay; TBST, Tris-buffered
saline/Tween 20; MAPK, mitogen-activated protein kinase; JNK, c-Jun N-terminal kinase; AP, activator protein; iNOS, inducible nitric-oxide
synthase; EMSA, electrophoretic mobility shift assay.
1237

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
1238
Park et al.
2002; Lim et al., 2000, 2001). In addition, the etiological role dition, it was found that the NF-␬B and ERK signaling path-
of microglial activation has been reported in other neurode- ways are involved in the anti-inflammatory effects of irisoli-
generative diseases (McGeer et al., 1988; Gao et al., 2003). done, which had the most potent anti-inflammatory effect of
Although the relationship between microglial activation and the six isoflavones tested. Considering that isoflavones and
neurodegenerative diseases is not completely understood, it their metabolites have minimal side effects in the body, the
is evident that microglial activation plays an important role inhibition of microglial activation by isoflavone metabolites
in the progression of several neurodegenerative diseases may be a good potential therapeutic modality for various
(Wyss-Coray and Mucke, 2002; Wyss-Coray, 2006). There- neurodegenerative diseases.
fore, the inhibition of microglial activation would be an effec-
tive therapeutic approach to alleviating the progression of
neurodegenerative diseases such as Alzheimer’s disease,
Materials and Methods
Parkinson’s disease, and multiple sclerosis.
Preparation of Three Isoflavones and Its Bacterial Metab-
olites. Glycitin, tectoridin, and kakkalide were isolated according to
the previous methods (Bae et al., 1999; Lee et al., 2000; Yamaki et
al., 2002; Han et al., 2003). The flowers of P. thunbergiana (500 g),
which were produced in Korea, were extracted with 2.5 liters of
boiling water, concentrated in a rotary evaporator, extracted three
times with ethyl acetate, and evaporated. The resulting extract (28 g)
was loaded on a silica-gel flash column chromatograph and eluted
with CHCl3:MeOH (20:1 3 4:1). We isolated glycitin (0.25 g), tecto-
ridin (0.3 g), and kakkalide (2.3 g). Each isolated compound (0.2 g)
was incubated with Bacteriodes spercoris HJ-15, a human intestinal
Isoflavones are biologically active compounds that are
found in a variety of plants, with relatively high levels being
found in soybean. Recently, a series of isoflavonoids including
kakkalide, tectoridin, and glycitin were isolated from the
flowers of Pueraria thunbergiana (Leguminoseae), which are
used in traditional Chinese medicine (Park et al., 1999).
Because most traditional medicines are administered orally,
their components inevitably come into contact with the in-
testinal microflora in the alimentary tract (Han et al., 2003).
These intestinal bacteria transform most of the components bacterium, in a final volume of 500 ml of anaerobic dilution medium
in an anaerobic glove box (Coy Laboratory Products Inc., Grass Lake,
MI), extracted three times with ethyl acetate, and evaporated. The
resulting extract (28 g) was loaded on a silica-gel flash column
chromatograph and eluted with CHCl3:MeOH (20:1 3 10:1). We
isolated their metabolites, glycitein (32 mg), tectorigenin (25 mg),
and irisolidone (36 mg): irisolidone [pale yellowish amorphous pow-
before they are absorbed through the gastrointestinal tract.
The human intestinal bacteria transform kakkalide, tectori-
din, and glycitin into irisolidone, tectorigenin, and glycitein,
respectively (Bae et al., 1999; Yamaki et al., 2002). The
transformed metabolites have more potent hepatoprotective
and anti-inflammatory activity than the glycoside form of
isoflavones (Yamaki et al., 2002; Han et al., 2003). This
suggests that kakkalide, tectoridin, and glycitin are prodrugs
that can be transformed into the active compounds by human
intestinal bacteria.
The isoflavones isolated from the rhizomes of P. thunber-
giana have been used in traditional medicine as antipyretics
and analgesics in treatment of the cold and whose flowers
have been used to treat diabetes mellitus, ethanol-induced
cell mortality, and hepatic injury (Kim, 1996). Recently,
there have been several reports showing the anti-inflamma-
tory effects of isoflavones in peripheral macrophages. Tecto-
rigenin and tectoridin inhibited the production of prostaglan-
din E2 and the expression of COX-2 in rat peritoneal
der, m.p. 189 to 190°C; IR ␯
(KBr), 3447, 2991, 1658, and 1033
max
cmϪ1; FAB-MS, 315 [M ϩ H]ϩ]; tectorigenin [pale yellowish amor-
phous powder, m.p. 230 to 233°C; IR ␯
(KBr), 3447, 2921, 1648,
max
and 1023 cmϪ1; FAB-MS, 303 [M ϩ H]ϩ]; and glycitein [pale yellow-
ish amorphous powder, m.p. 178 to 180°C; IR ␯max (KBr), 3472, 1635,
and 1178 cmϪ1; FAB-MS, 315 [MϩH]ϩ; FAB-MS, 290 [MϩH]ϩ]. The
chemical structures of isoflavones and their metabolites are shown in
Fig. 1.
Reagents. LPS (Escherichia coli serotype 055:B5) was obtained
from Sigma-Aldrich (St. Louis, MO). Two kinds of ERK inhibitors,
PD98059 and U0126, and two kinds of NF-␬B inhibitors, pyrrolidone
dithiocarbamate (PDTC) and MG132, were purchased from Calbio-
chem (La Jolla, CA). All other chemicals were obtained from Sigma-
Aldrich, unless stated otherwise.
Microglial Cell Cultures and Cell Viability. The immortalized
macrophages (Kim et al., 1999). It was reported that irisoli- murine BV2 microglial cell line was grown and maintained in Dul-
becco’s modified Eagle’s medium supplemented with 10% heat-inac-
tivated fetal bovine serum, streptomycin (10 ␮g/ml), and penicillin
(10 U/ml) at 37°C. Cultures of primary microglial cells were estab-
lished based on the differential adherence of cells harvested from
fetus rat cortex. The methods were modified from Frei et al. (1987).
Mixed cell cultures were prepared from postnatal 1-day rat cerebral
cortices. Cortices were dissociated by passing through a 130-␮m
nylon mesh and plated at 2 ϫ 105 cells/cm2 in T75 Falcon flasks.
Cultures were fed every 3 to 4 days with modified Eagle’s medium
(MEM) supplemented with 10% fetal calf serum (FCS). On days 12 to
done, tectorigenin, genistein, and glycitein suppress 12-O-
tetradecanoylphorbol-13-acetate-induced prostaglandin E2
production (Yamaki et al., 2002). Meanwhile, glycitein was
shown to inhibit LPS-induced NO production in Raw264.7
macrophage cells (Sheu et al., 2001). These results have
highlighted the therapeutic potential of these isoflavones in
various inflammatory diseases.
Despite the anti-inflammatory effects in peripheral macro-
phage cells, the effects of the six isoflavones in microglial
activation have not been reported. Moreover, the detailed 13, the culture plate was shaken on a rotatory shaker at 200 rpm for
30 min. The suspended cells were plated on 24- or 48-well culture
plates. After 1-h incubation at 37°C, the medium containing sus-
pended cells was discarded, and adherent cells were further incu-
bated with 1% FCS-supplemented MEM for future experiments. The
homogeneity of the culture was determined by immunostaining for
Mac-1 cell surface antigen expression and was routinely found to be
higher than 90%.
Cell viability was determined in microglial cells by MTT reduction
assay. In brief, cells were added to MTT (5 mg/ml in phosphate-
buffered saline) and incubated at 37°C for 2 to 3 h. The resulting
molecular/signaling mechanisms underlying the anti-inflam-
matory effects of these isoflavones are not completely under-
stood. Therefore, this study examined whether or not the
three isoflavones isolated from the flowers of P. thunbergiana
and their bacterial metabolites suppress microglial activa-
tion. It was found that the isoflavone metabolites (irisolidone,
tectorigenin, glycitein) strongly suppressed the proinflamma-
tory cytokines and NO production in LPS-stimulated micro-
glial cells, whereas the glycoside isoflavones (kakkalide,
tectoridin, glycitin) showed no significant inhibition. In ad- dark blue crystals were dissolved with an equal volume of isopropa-

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Anti-Inflammatory Mechanisms of Isoflavone Metabolites
1239
Fig. 1. The chemical structures of isoflavones used in this study. Glycosylated isoflavones and their aglycones (bacterial metabolites) are shown.
nol containing 40 mM HCl. Absorbance was determined at a test the pro- and mature form of TNF-␣ (Santa Cruz Biotechnology, Inc.,
wavelength of 570 nm and a reference wavelength of 630 nm using a
Microplate reader (SpectraMax 340pc; Molecular Devices, Sunny-
vale, CA).
Santa Cruz, CA) was used. After thoroughly washing with TBST,
horseradish peroxidase-conjugated secondary antibodies (New En-
gland Biolabs, Beverly, MA; 1:3000 dilution in TBST) were applied,
and the blots were developed using an enhanced chemiluminescence
detection kit (GE Healthcare, Little Chalfont, Buckinghamshire,
UK).
Measurement of Cytokine and Nitrite Levels. Microglial cells
(1 ϫ 105 cells per well in a 24-well plate) were pretreated with
isoflavones for 30 min and stimulated with LPS (0.1 ␮g/ml) in the
presence of serum. The supernatants of the cultured microglia were
collected 6 or 24 h after LPS stimulation, and the concentrations of
TNF-␣ and IL-1␤ were measured by enzyme-linked immunosorbent
assay (ELISA) using monoclonal antibodies and the procedure rec-
ommended by the supplier (PharMingen, San Diego, CA). The serum
in the media did not interfere with the assay. Accumulated nitrite
was measured in the cell supernatant by the Griess reaction (Green
et al., 1982). The conditions of cell culture and treatment were same
with those in ELISA. In brief, 50-␮l aliquots of cell supernatant from
each well were mixed with 100 ␮l of Griess reagent (mixing equal
volumes of 0.1% naphthylethylenediamine dihydrochloride and 1%
sulfanilamide in 5% phosphoric acid) in a 96-well microtiter plate,
and absorbance was read at 540 nm using a plate reader (Molecular
Devices). Sodium nitrite, diluted in culture media at concentrations
ranging from 10 to 100 ␮M, was used to prepare a standard curve.
RNase Protection Assay. BV2 cells (4 ϫ 105 cells on a 35-mm
dish) were treated with LPS in the presence or absence of isoflavone
metabolites, and total RNA was extracted with TRIzol reagent
(Sigma-Aldrich) according to the manufacturer’s protocol. RNase
protection assay was performed using a Riboquant multiprobe
RNase protection assay kit (PharMingen). Protected fragments were
resolved on 6% urea-polyacrylamide gel, and autography was per-
formed after exposing the dried gel to X-ray film.
Nuclear Extract Preparation and Electrophoretic Mobility
Shift Assay. Nuclear extracts from treated microglia were prepared
as described previously (Woo et al., 2003). The double-stranded DNA
oligonucleotides containing the NF-␬B or AP-1 consensus sequences
(Promega) were end-labeled using T4 polynucleotide kinase (New
England Biolabs) in the presence of [␥-32P]ATP. Five micrograms of
the nuclear proteins was incubated with 32P-labeled NF-␬B or AP-1
probe on ice for 30 min and resolved on a 5% acrylamide gel as
described previously (Woo et al., 2003). For the supershift assay,
antibodies to the p65 or p50 subunits of NF-␬B (Santa Cruz Biotech-
nology, Inc.) were coincubated with the protein in the reaction mix-
ture for 30 min at 4°C before adding the radiolabeled probe. For
competition binding assays, binding reaction reagents and nuclear
extracts were mixed with nonradioactive oligonucleotides in molar
excess and incubated before adding 32P-labeled probe.
Transient Transfection and Luciferase Assay. Transfection
of the NF-␬B or AP-1 reporter gene into BV2 cells was performed
using Geneporter 2 transfection reagent (Gene Therapy Systems
Inc., San Diego, CA). BV2 cells (1 ϫ 105 per well in a 12-well plate)
were transfected with 1 ␮g of the reporter construct mixed with
Geneporter. After 48 h, cells were harvested, and a luciferase assay
was performed as described previously (Woo et al., 2003). To deter-
mine the effect of isoflavones on LPS-induced NF-␬B or AP-1 activ-
ity, cells were pretreated with the agents for 30 min before treating
with LPS and incubated for 6 h before harvesting cells for luciferase
assay. Transfection assays were performed three times in duplicate.
Statistical Analysis. The SAS program (SAS Institute, Cary,
NC) was used for statistical analysis. The data are expressed as the
mean Ϯ S.E.M. and analyzed for statistical significance using anal-
ysis of variance, followed by Scheffe’s test for multiple comparison. A
Western Blot. Cell extracts were prepared as described previ-
ously (Woo et al., 2003), and 50 ␮g of proteins was separated by
SDS-polyacrylamide gel electrophoresis and transferred to polyvi-
nylidene difluoride membranes. The membranes were blocked with
5% bovine serum albumin in 10 mM Tris-HCl containing 150 mM
NaCl and 0.5% Tween 20 (TBST) and then incubated with primary
antibodies (1:1000) that recognize the phospho- or total forms of
ERK1/2, p38 MAPK, and JNK (Cell Signaling Technology Inc., Bev-
erly, MA). For Western blot of TNF-␣, antibody that recognizes both p value Ͻ 0.05 was considered significant.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
1240
Park et al.
metabolites at 50 ␮M significantly inhibited iNOS, TNF-␣,
and IL-1␤ mRNA expression. Despite the robust inhibition of
TNF-␣ protein production by 50 ␮M of the isoflavone metab-
olites, the same concentration of metabolites decreased the
TNF-␣ mRNA levels by only 20 to 30%. Therefore, these
results suggest that three isoflavone metabolites regulate the
expression of iNOS and IL-1␤ at the transcriptional level,
whereas they regulate the expression of TNF-␣ at the tran-
scriptional and/or post-transcriptional level. In support of
this hypothesis, Western blot analysis revealed that the
isoflavone metabolites significantly repressed both the pro-
and mature form TNF-␣ at the protein level in the order of
irisolidone Ͼ glycitein Ͼ tectorigenin, as shown in ELISA
data (Fig. 3C). Based on these results, the molecular mech-
anism of irisolidone, which showed the most potent anti-
inflammatory activity among three isoflavone metabolites
tested, was further examined.
Irisolidone Inhibits the DNA Binding and Transcrip-
tional Activities of NF-␬B and AP-1 in LPS-Stimulated
Microglial Cells. NF-␬B is an important upstream modula-
tor of cytokine and iNOS expression in microglia (Pahl,
1999), and the anti-inflammatory actions of a number of
cytokines and chemicals have been explained in terms of the
inhibition of NF-␬B activity. Therefore, the involvement of
NF-␬B activity in irisolidone-induced suppression of NO and
cytokines was examined. As shown in Fig. 4A, the stimula-
tion of BV2 cells with LPS resulted in strong NF-␬B binding,
which was inhibited significantly by irisolidone. The com-
pound also affected NF-␬B-mediated transcription (Fig. 4B).
The NF-␬B transcriptional activity was assayed by transfect-
ing the BV2 cells with a plasmid containing three NF-␬B
binding sites and a luciferase reporter gene. The luciferase
assay showed that irisolidone repressed the LPS-stimulated
Results
Effect of the Isoflavones on NO and Cytokine Pro-
duction in LPS-Stimulated Microglial Cells. The BV2
microglial cells were stimulated with LPS (100 ng/ml) in the
presence or absence of the isoflavones isolated from the flow-
ers of P. thunbergiana to determine whether they reduce the
level of NO, TNF-␣, and IL-1␤ production, which play impor-
tant roles in various inflammatory diseases. The stimulation
of BV2 cells with LPS led to a significant increase in the NO,
TNF-␣, and IL-1␤ levels in the cell-conditioned media after
24 h. Pretreatment of BV2 cells with the isoflavone metabo-
lites (irisolidone, tectorigenin, glycitein) significantly inhib-
ited the LPS-induced NO, TNF-␣, and IL-1␤ production in a
dose-dependent manner (Fig. 2A). The potency of the inhibi-
tion of NO and TNF-␣ was in order of irisolidone Ͼ gly-
citein Ͼ tectorigenin, and the inhibition of IL-1␤ was in the
order of glycitein Ͼ irisolidone Ͼ tectorigenin. In contrast,
their proforms (glycoside isoflavones) showed no significant
inhibition. The isoflavone metabolite-induced decrease in
NO, TNF-␣, and IL-1␤ was also observed in the rat primary
microglia cultures (Fig. 2B). This suggests that the anti-
inflammatory effect of isoflavone metabolites is not limited to
a particular microglial cell line. To exclude the possibility
that the decrease in the NO and cytokine levels was simply
due to cell death, the cell viability was tested at various
concentrations of isoflavones. The MTT assay showed that
the isoflavones did not have any cytotoxicity at the concen-
trations (5–50 ␮M) used in this study in both the BV2 and
primary microglial cells for at least 48 h (data not shown).
Isoflavone Metabolites Suppress mRNA Expressions
of iNOS, TNF-␣, and IL-1␤. RNase protection assay was
performed to determine whether the isoflavone metabolites
regulate iNOS and cytokine expression at the transcriptional
level. As shown in Fig. 3A, the three types of isoflavone NF-␬B transcriptional activation in a dose-dependent man-
Fig. 2. Effect of isoflavones on NO and
cytokine production in LPS-stimu-
lated BV2 cells (A) and primary mi-
croglia (B). BV2 microglial cells (1 ϫ
105 cells per well in a 24-well plate)
were incubated with 0.1 ␮g/ml LPS
and various concentrations of isofla-
vones in DMEM containing 10% fetal
bovine serum for 24 h. Rat primary
microglia were incubated with 10
ng/ml LPS and isoflavones in 1% FCS-
supplemented MEM. The amounts of
NO in the supernatants were mea-
sured using Griess reagent and
TNF-␣ and IL-1␤ by ELISA. The data
are expressed as the mean Ϯ S.E.M. of
four independent experiments. ء, p Ͻ
0.05; ءء, p Ͻ 0.01; Scheffe’s test; sig-
nificantly different from the value in
cells treated with LPS in the absence
of isoflavones. The vehicle (0.2% di-
methyl sulfoxide) where isoflavones
were dissolved did not affect either
basal or LPS-induced NO and cyto-
kine levels.