Structural requirements of sesquiterpene lactones to inhibit LPS-induced nitric oxide synthesis in RAW 264.7 macrophages

Article abstract of DOI:10.1016/S0968-0896(00)00202-9

Some sesquiterpene lactones were recently demonstrated to inhibit inducible nitric oxide synthase (iNOS)-dependent nitric oxide (NO) synthesis. The primary objective of the present study was, therefore, to find evidence for structural requirements of sesquiterpene lactones regarding their capability to inhibit iNOS-dependent NO synthesis. Sesquiterpene lactones 1-11 were examined for their influence on nitrite accumulation in cell culture supernatants of LPS-induced RAW 264.7 macrophages. Except the taraxinic acid β-D-glucopyranosylester 8 all compounds showed a dose-dependent inhibition of nitrite accumulation in cell culture supernatants with IC₅₀ values ranging from 0.5 to 36.8 μM. High activity seemed to be dependent on an α-methylene-γ-lactone functionality. Cytotoxicity and the ability to inhibit activation of transcription factor NF-κB are further biological activities of sesquiterpene lactones. The second point of interest was, therefore, whether the structural requirements of sesquiterpene lactones for these activities may differ or be the same for those needed to inhibit iNOS-dependent NO synthesis. Using concentrations of 1-11 required to inhibit NO synthesis cell viability was determined and NF-κB binding activity was measured by gel-shift experiments. Interestingly, compounds almost equally effective in inhibiting nitrite accumulation did not show the same cytotoxic potential, and most sesquiterpene lactones inhibited nitrite accumulation at concentrations where inhibition of NF-κB activation was not significant. These results suggest that different biological activities of sesquiterpene lactones have different structural requirements. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00202-9

Bioorganic & Medicinal Chemistry 8 (2000) 2747±2753
Structural Requirements of Sesquiterpene Lactones to Inhibit
LPS-Induced Nitric Oxide Synthesis in RAW 264.7 Macrophages
Verena M. Dirsch, a,* Hermann Stuppner, b Ernst P. Ellmerer-Muller c
and Angelika M. Vollmar a
aInstitute of Pharmacy, Center of Drug Research, University of Munich, D-81377 Munich, Germany
bInstitute of Pharmacy, University of Innsbruck, A-6020 Innsbruck, Austria
cInstitute of Organic Chemistry, University of Innsbruck, A-6020 Innsbruck, Austria
Received 9 March 2000; accepted 27 July 2000
AbstractÐSome sesquiterpene lactones were recently demonstrated to inhibit inducible nitric oxide synthase (iNOS)-dependent
nitric oxide (NO) synthesis. The primary objective of the present study was, therefore, to ®nd evidence for structural requirements
of sesquiterpene lactones regarding their capability to inhibit iNOS-dependent NO synthesis. Sesquiterpene lactones 1±11 were
examined for their in¯uence on nitrite accumulation in cell culture supernatants of LPS-induced RAW 264.7 macrophages. Except
the taraxinic acid b-d-glucopyranosylester 8 all compounds showed a dose-dependent inhibition of nitrite accumulation in cell
culture supernatants with IC50 values ranging from 0.5 to 36.8 mM. High activity seemed to be dependent on an a-methylene-g-
lactone functionality. Cytotoxicity and the ability to inhibit activation of transcription factor NF-kB are further biological activities
of sesquiterpene lactones. The second point of interest was, therefore, whether the structural requirements of sesquiterpene lactones
for these activities may di€er or be the same for those needed to inhibit iNOS-dependent NO synthesis. Using concentrations of 1±
11 required to inhibit NO synthesis cell viability was determined and NF-kB binding activity was measured by gel-shift experiments.
Interestingly, compounds almost equally e€ective in inhibiting nitrite accumulation did not show the same cytotoxic potential, and
most sesquiterpene lactones inhibited nitrite accumulation at concentrations where inhibition of NF-kB activation was not sig-
ni®cant. These results suggest that di€erent biological activities of sesquiterpene lactones have di€erent structural requirements.
# 2000 Elsevier Science Ltd. All rights reserved.
Introduction
systems.9 11 iNOS is an enzyme which is induced in dif-
ferent cell types by pro-in¯ammatory stimuli.12 Exces-
sive production of nitric oxide (NO) by this enzyme is
considered as a promoter of tissue injury in in¯amma-
tion. iNOS is, therefore, an important target involved in
in¯ammatory and also immunoregulatory processes.13
The ability of sesquiterpene lactones to modulate iNOS
expression is suggested to contribute to the anti-in¯am-
matory properties of these compounds.9
Sesquiterpene lactones are biologically active natural
products found almost exclusively in species of the
Compositae (Asteraceae) family. Common structural
features of this class of compounds are a terpenoid C-15
skeleton and a g-lactone moiety. Other structural ele-
ments can vary considerably. Depending on the basic
skeleton, di€erent subgroups are distinguished, e.g., ger-
macranolides, eudesmanolides, guaianolides and pseu-
doguaianolides. Pharmacological activities described for
sesquiterpene lactones include antimicrobial,1 antiviral,2
cytotoxic or antitumor,1,3 and anti-in¯ammatory prop-
erties.4 8 Recently, several sesquiterpene lactones, i.e.,
parthenolide, isohelenin, dehydrocostus lactone, and
yomogin, were shown to inhibit the expression of
inducible nitric oxide synthase (iNOS) in various cell
In this context, studies aimed at elucidating the struc-
tural requirements of sesquiterpene lactones regarding
their potential to inhibit iNOS-dependent NO synthesis
are important to ®nd new potent inhibitors. The pri-
mary objective of this study was, therefore, to investi-
gate sesquiterpene lactones from di€erent structural
subgroups for their potential to inhibit iNOS-dependent
NO synthesis in order to ®nd structural requirements
for this biological activity.
So far, the best characterized biological activities of
sesquiterpene lactones are their cytotoxicity1,3,14,15 and
*Corresponding author. Tel.: +89-2180-7161; fax: +89-2180-7173; e-
mail: verena.dirsch@cup.uni-muenchen.de
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00202-9

2748
V. M. Dirsch et al. / Bioorg. Med. Chem. 8 (2000) 2747±2753
the ability of sesquiterpene lactones to inhibit the acti-
vation of the transcription factor nuclear factor kappaB
(NF-kB).5 9 Both activities are of interest in the context
of the present study: in vitro cytotoxicity may simulate
inhibition of NO synthesis and is undesirable from a
pharmacological point of view. Activation of NF-kB
was shown to be necessary for iNOS expression.16
Moreover, Wong et al. provided evidence that sesqui-
terpene lactones inhibit iNOS expression via inhibition
of NF-kB.9 Therefore, our second point of interest was
to learn whether the structural requirements of sesqui-
terpene lactones needed for the inhibition of nitric oxide
synthesis correlate with those necessary for their cyto-
toxic potential and their ability to inhibit activation of
NF-kB.
For these studies RAW 264.7 macrophages activated by
bacterial endotoxic lipopolysaccharide (LPS) were cho-
sen since induction of iNOS was shown to be dependent
on activation of NF-kB in this cell system.16 Sesqui-
terpene lactones 1±11 (Chart 1) were selected from dif-
ferent structural subgroups. First, the in¯uence of 1±11
on iNOS-dependent NO synthesis was determined by
measuring nitrite accumulation in cell culture super-
Chart 1. Sesquiterpene lactones 1±11.

V. M. Dirsch et al. / Bioorg. Med. Chem. 8 (2000) 2747±2753
2749
natants. Second, using the determined e€ective con-
centrations of 1±11 the in¯uence on cell viability and
NF-kB activation was measured.
iNOS-dependent NO synthesis and their number of
reactive a,b-unsaturated carbonyl structure elements.
Table 1 shows that there is no direct correlation
between the number of potentially active centers of ses-
quiterpene lactones and their NO inhibitory activity.
Even within the group of pseudoguaianolide-type ses-
quiterpene lactones ambrosin 1 (IC50 3.2 mM, two active
centers) shows the same activity as bipinnatin 2 (IC50
3.3 mM, one active center). However, the combination
of a lactone ring with a conjugated exomethylene group
(a-methylene-g-lactone) seems to be important, since
glaucolide A 7, the only sesquiterpene lactone missing
the exomethylene group in the lactone ring, showed the
weakest activity (IC50 36.8 mM) of all non-glycosylated
sesquiterpene lactones investigated (Table 1). To sup-
port the idea that the a-methylene-g-lactone moiety
might be essential for the NO inhibitory activity, we
hydrogenated the double bonds in alantolactone 6.
Alantolactone 6 was chosen since it was the most active
constituent bearing as the only reactive functional ele-
ment the a-methylene-g-lactone. Indeed, we found that
the obtained mixture (the dihydroalantolactone lacking
the exocyclic methylene group and the tetrahydro-alan-
tolactone in a ratio of 45:55) showed no signi®cant
activity up to 50 mM (data not shown).
Results
Inhibition of LPS-induced nitrite accumulation in RAW
264.7 cells by sesquiterpene lactones 1±11. Structural
requirements
In order to investigate whether sesquiterpene lactones
1±11 are able to inhibit inducible nitric oxide synthase
(iNOS)-dependent nitric oxide (NO) synthesis, we deter-
mined nitrite accumulation (20 h) in cell culture super-
natants of lipopolysaccharide (LPS)-activated (1 mg/mL)
and sesquiterpene lactone-treated RAW 264.7 macro-
phages seeded in 96-well plates. Maximum concentra-
tion of sesquiterpene lactones 1±11 employed was 100
mM. Table 1 shows that indeed all non-glycosylated
sesquiterpene lactones were able to inhibit NO synthesis
with IC50 values far below 100 mM. The highest activity
(IC50<1 mM) was found for hymenin 4 (IC50 0.5 mM),
alantolactone 6 (IC50 0.6 mM) and helenalin 3 (IC50 0.9
mM), followed by compounds with IC50 values between
1 and 10 mM, namely ambrosin 1 (IC50 3.2 mM), bipin-
natin 2 (IC50 3.3 mM), xanthatin 11 (IC50 5.0 mM), psi-
lotropin 10 (IC50 6.2 mM) and psilostachyin 9 (IC50 8.7
mM). IC50 values >10 mM were determined for gaillar-
din 5 (IC50 12.2 mM), glaucolide A 7 (IC50 36.8 mM) and
taraxinic acid b-d-glucopyranosylester 8 (IC50>100
mM). All e€ects were dose-dependent (data not shown).
Glycosylation seems to negatively modulate activity, as
seen by taraxinic acid-b-d-glucopyranosylester 8 (Table
1).
In¯uence of 1±11 on cell viability in comparison to
inhibition of nitrite accumulation
The biological activity of sesquiterpene lactones was
suggested to be mediated by the reaction of a-methyl-
ene-g-lactones and other conjugated systems with sulf-
hydryl groups. For various activities the a-methylene-g-
lactone moiety seems to be the most important func-
tional group. Other a,b-unsaturated carbonyl structure
elements were reported to enhance these activities.6,14,15
We, therefore, tried to ®nd a correlation between the
potential of sesquiterpene lactones 1±11 to inhibit
Viability of cells was determined after removing cell
culture supernatants for nitrite measurement, i.e., after
an incubation time of 20 h. As seen in Table 1, no ses-
quiterpene lactone impaired cell viability at the con-
centration determined as IC50 value regarding inhibition
of nitrite accumulation. However, all sesquiterpene lac-
tones e€ective in inhibiting nitrite accumulation showed
cytotoxicity at some higher concentrations. Interest-
ingly, compounds almost equally e€ective in inhibiting
Table 1. IC50 values obtained from measuring nitrite accumulation, cell viability and structural functionality of 1±11
Compound
IC50 value
(mM)
Unimpaired cell
viability observed
until (mM)
Mono- or bi-
functional
a-Methylene-
g-lactone
Sugar
moiety
Hymenin 4
0.5
0.6
0.9
3.2
3.3
5.0
6.2
8.7
10
2
2
bi (mla; cpb)
mono (mL)
bi (ml; cp)
bi (ml; cp)
mono (mL)
bi (ml; ucc)
mono (mL)
mono (mL)
mono (mL)
bi (uc; epd)
bi (ml; uc)
@
@
@
@
@
@
@
@
@
Alantolactone 6
Helenalin 3
Ambrosin 1
5
Bipinnatin 2
50
25
50
50
50
75
100
Xanthatin 11
Psilotropin 10
Psilostachyin 9
Gaillardin 5
Glaucolide A 7
Taraxinic acid-b-d-gluco-
pyranosylester 8
12.2
36.8
>100
@
@
aa-Methylene-g-lactone.
ba,b-Unsubstituted cyclopentenone.
ca,b-Unsaturated carbonyl structure.
dEpoxide.

2750
V. M. Dirsch et al. / Bioorg. Med. Chem. 8 (2000) 2747±2753
nitrite accumulation, did not show the same cytotoxic
potential, e.g., helenalin 3 (IC50 0.9 mM) showed quite
profound cytotoxicity at a concentration as low as 5 mM
(viability 53%, n=4; SEM: 0.023), whereas the even
more e€ective hymenin 4 (IC50 0.5 mM) did not show
cytotoxicity up to 10 mM. Similarly, ambrosin 1 (IC50
3.2 mM) showed ®rst signs of cytotoxicity at a con-
centration of 10 mM (viability 91%, n=3; SEM: 0.024),
whereas bipinnatin 2 (IC50 3.3 mM) showed no cyto-
toxicity at all until 50 mM.
nitrite accumulation. As can be seen in Figure 2, most
sesquiterpene lactones applied at concentrations sig-
ni®cantly inhibiting iNOS-dependent NO synthesis do
not show a signi®cant inhibition of NF-kB activation.
Only psilotropin 10 and gaillardin 5 are able to reduce
iNOS-dependent NO synthesis and NF-kB activation to
a similar extent.
Interestingly, PDTC (50 mM), which was shown to act
not only as inhibitor of NF-kB activation but also as
e€ective scavenger of NO formation,17 inhibited nitrite
accumulation more markedly ( 84.2%, n=8, SEM:
2.183) than NF-kB activation ( 54.5%, n=8, SEM:
9.272) in LPS-activated RAW 264.7 cells (Fig. 2).
In¯uence of 1±11 on NF-ꢀB binding activity in
comparison to inhibition of nitrite accumulation
All sesquiterpene lactones except the less e€ective gly-
cosylated 8 were examined for their in¯uence on NF-kB
binding activity in LPS-activated RAW 264.7 cells by
the electrophoretic mobility shift assay (EMSA). Con-
centrations employed in this assay were within the
determined non-toxic range and above the ascertained
IC50 values regarding nitrite accumulation (Table 1). To
obtain enough nuclear protein for EMSA experiments
we had to switch from cell culturing in 96-well plates to
6-well plates. Since di€erent cell culture conditions may
a€ect experimental results, i.e., results obtained in 6-
well plates might di€er from those obtained in 96-well
plates, all EMSA experiments were set up in parallel for
measuring nitrite accumulation. As a positive control
we used PDTC (pyrrolidinedithiocarbamate), a known
inhibitor of NF-kB activation and iNOS induction. A
representative gel-shift experiment is shown in Figure
1. Experiments, each performed at least three times,
were quanti®ed by densitometry and the results com-
pared to values obtained in parallel from measuring
Discussion
The present study demonstrates that (i) all sesquiterpene
lactones tested, except the glycosylated 8, are able to
inhibit iNOS-dependent nitrite accumulation dose-
dependently with IC50 values ranging from 0.5 mM to
36.8 mM; (ii) a high inhibitory activity correlated with
the existence of an a-methylene-g-lactone moiety but
not necessarily with a bi-functionality of the respective
compound; (iii) the inhibitory activities were not due to
cytotoxicity. The cytotoxic potential observed at higher
concentrations did not necessarily correlate with the
ability to inhibit NO synthesis; (iv) most sesquiterpene
lactones inhibited iNOS-dependent nitrite accumulation
at concentrations where inhibition of NF-kB activation
was not signi®cant. The potency of sesquiterpene lactones
to inhibit activation of NF-kB did not show a general
correlation to the ability to inhibit NO synthesis.
The report that some sesquiterpene lactones inhibit
iNOS-dependent nitric oxide synthesis9 11 raised the
question whether the ability to inhibit iNOS-dependent
NO synthesis may be a common property of sesqui-
terpene lactones and which structural requirements
might be necessary for this activity. We demonstrated
that, indeed, a broad range of structural diverse con-
stituents 1±11 were able to block LPS-induced NO
synthesis in RAW 264.7 cells with di€erent ecacy.
Comparisons regarding structure±activity relationships
showed that all sesquiterpene lactones with comparatively
high activities (IC50<15 mM) contained a conjugated
exomethylene group in their lactone ring (a-methylene-
g-lactone). Glaucolide A 7, the only sesquiterpene lac-
tone lacking this a-methylene-g-lactone group, was
despite other potentially reactive elements (a,b-unsatu-
rated carbonyl groups, epoxide group) less active than
all other non-glycosylated sesquiterpene lactones inves-
tigated. The functional signi®cance of the a-methylene-
g-lactone moiety was supported testing the mixture of
diastereomers obtained upon hydrogenation of alanto-
lactone 6. Alantolactone bearing an a-methylene-g-lac-
tone was found to be the most active mono-functional
sesquiterpene lactone (IC50 0.6 mM). After hydrogena-
tion, the obtained mixture showed no signi®cant e€ect
on nitrite accumulation up to 50 mM. The observed
impact of an a-methylene-g-lactone group is in accor-
dance with studies examining other biological activities
Figure 1. Representative gel-shift experiment showing the e€ect of
sesquiterpene lactones ambrosin 1 (5 mM), bipinnatin 2 (5 mM), and
gaillardin 5 (20 mM) as well as the positive control PDTC (50 mM) on
the NF-kB binding activity in nuclear extracts of RAW 264.7 cells
activated with LPS (1 mg/mL) for 1 h. Equal amounts of nuclear
extracts (10 mg) were subjected to EMSA as described in the Experi-
mental. Lane (1) untreated cells; lane (2) cells activated with LPS; lane
(3) cells pretreated with PDTC (50 mM) for 2 h and then activated with
LPS; lanes (4)±(6) cells pretreated with ambrosin 1 (5 mM, lane 4),
bipinnatin 2 (5 mM, lane 5), gaillardin 5 (20 mM, lane 6) for 1 h and
then activated with LPS; lane (7) nuclear extract of LPS-activated cells
incubated with a 100-fold excess of unlabeled NF-kB consensus oli-
gonucleotide (unlabeled competitor); lane (8) nuclear extract of LPS-
activated cells incubated with a 100-fold excess of unlabeled AP-2
consensus oligonucleotide (unlabeled non-competitor).

V. M. Dirsch et al. / Bioorg. Med. Chem. 8 (2000) 2747±2753
2751
Figure 2. Comparison of the ability of sesquiterpene lactones 1±7 and 9±11 to inhibit nitrite accumulation in cell culture supernatants of LPS (1 mg/
mL)-stimulated RAW 264.7 macrophages and NF-kB binding activity in nuclear extracts of RAW 264.7 cells activated with LPS (1 mg/mL) for 1 h.
Values are given in % of the nitrite concentration and NF-kB binding activity detected in cell culture supernatants or nuclear extracts, respectively,
of cells activated with LPS (100%). Employed concentrations: PDTC 50 mM, hymenin 2 mM, alantolactone 2 mM, helenalin 3 mM, ambrosin 5 mM,
bipinnatin 5 mM, xanthatin 10 mM, psilotropin 10 mM, psilostachyin 15 mM, gaillardin 20 mM, glaucolide A 50 mM. Bars represent mean (ÆSEM) of
at least three independent experiments. Nitrite measurement was performed in triplicate. *P<0.05, **P<0.01 (ANOVA/Dunnett).
of sesquiterpene lactones: Kupchan et al. reported that
the presence of an a-methylene-g-lactone is essential for
signi®cant cytotoxic activity of sesquiterpene lactones.14
Accordingly, investigations regarding the structural
requirements for the inhibitory activity of sesquiterpene
lactones on NF-kB activation found that compounds
lacking this moiety display no inhibitory e€ect on acti-
vation of NF-kB.8 On the other hand, other functional
elements such as an a,b-unsaturated carbonyl structure
or an epoxide group were suggested to enhance biologi-
cal activity of sesquiterpene lactones.5,6,14 Our results do
not ®t into this notion: as Table 1 demonstrates there was
no obvious correlation between the IC50 values and a
mono- or bi-functionality. For example, ambrosin 1 con-
taining an a,b-unsubstituted cyclopentenone in addition
to the a-methylene-g-lactone moiety inhibited nitrite
accumulation equally to bipinnatin 2 with an a-methyl-
ene-g-lactone only.
therefore, performed our own experiments regarding these
two well documented e€ects of sesquiterpene lactones
comparing inhibition of nitric oxide synthesis, cytotoxi-
city, and inhibition of NF-kB in the same cell system.
Interestingly, we found that compounds with similar
ecacy regarding inhibition of nitrite accumulation did
not necessarily show the same cytotoxic activity when
applied at the same concentration. For instance, hyme-
nin 4 and bipinnatin 2 were found to be less cytotoxic
than helenalin 3 and ambrosin 1, respectively. Thus,
regarding a potential therapeutic use, hymenin and
bipinnatin are preferable. The di€erent cytotoxic poten-
tials of bipinnatin 2 and ambrosin 1 may be explained by
an additional a,b-unsubstituted cyclopentenone in the
ambrosin structure 1 compared to bipinnatin 2 with an
a-methylene-g-lactone element only.14
Measuring activation of NF-kB by gel-shift experi-
ments, we found that most sesquiterpene lactones were
more e€ective in inhibiting iNOS-dependent NO synth-
esis than NF-kB activation. These data suggest that
there might be an additional mechanism, other than the
sole inhibition of NF-kB, involved in the process lead-
ing to sesquiterpene lactone mediated reduction of NO
synthesis. The antioxidant PDTC, which also inhibited
nitrite accumulation more markedly than NF-kB activa-
tion (Fig. 2), was shown to act not only as inhibitor of
NF-kB activation but also as e€ective scavenger of NO
formation.18 This mechanism cannot be excluded for
sesquiterpene lactones. Conceivable additional mechan-
isms of action for sesquiterpene lactones might include
an inhibition of the enzymatic activity of iNOS or
interferences on other levels of iNOS regulation.
As observed previously studying glycosidic and non-
glycosidic hypocretenolides,17 a lipophilic character
seems to be advantageous for high inhibitory activity in
vitro. The fact that the taraxinic acid b-d-glucopyr-
anosylester 8 exerted activity only higher than 100 mM,
and the lipophilic alantolactone 6 was, despite its mono-
functionality, one of the most active compounds is in
line with this assumption.
The data presented here show that the structural
requirements of sesquiterpene lactones for an e€ective
inhibition of iNOS-dependent NO synthesis do not
correspond to those reported in the literature for other
biological activities, i.e., especially their cytotoxicity and
the ability to inhibit activation of NF-kB.5,6,14 We,

2752
V. M. Dirsch et al. / Bioorg. Med. Chem. 8 (2000) 2747±2753
Furthermore, comparing data from nitrite measurement
and gel-shift assays, we found that there was no general
correlation for the potency of sesquiterpene lactones to
inhibit NO synthesis and activation of NF-kB (Table 1),
supporting the idea that the structural determinants for
both activities might not be completely identical.
Cell culture
RAW 264.7 cells obtained from the American Type
Culture Collection (ATCC, TIB 71, Maryland, USA)
were cultured in Dulbecco's Modi®ed Essential Medium
with 4 mM l-glutamine and 4.5 g/l glucose (DMEM,
endotoxin level <0.005 EU/mL, Bio Whittaker, Bio-
products, Heidelberg, Germany), supplemented with
10% heat-inactivated fetal calf serum (Gibco/BRL Life
Technologies, Eggenstein, Germany). Cells were main-
tained at 37 ꢀC, 5% CO2 and used for experiments
between passages 5 and 20. In general, con¯uent cells
were stimulated with 1 mg/mL LPS (E. coli, Serotype
055:B5 Sigma, Deisenhofen, Germany). Sesquiterpene
lactones were dissolved in 10% dimethyl sulfoxide
(DMSO) at a concentration of 10 mM and further
diluted in phosphate-bu€ered saline (PBS). Final con-
centrations of DMSO in cell culture supernatants never
exceeded 0.1%, which was shown not to interfere with
the used assays. Pyrrolidinedithiocarbamate (50 mM;
PDTC; Calbiochem, Bad Soden, Germany), a known
inhibitor of NF-kB activation and iNOS induction, was
used as positive control.
Conclusion
We conclude that the most important structural
requirement for inhibition of iNOS-dependent NO
synthesis seems to be the a-methylene-g-lactone moiety.
Additional other functional elements like, e.g., an a,b-
unsubstituted cyclopentenone seem to be less important.
Data from the literature as well as our experiments
support the idea that di€erent biological activities of
sesquiterpene lactones seem to have di€erent structural
requirements. However, the a-methylene-g-lactone moi-
ety seems to play a crucial role for most activities
described so far. These data may provide a chance to
®nd compounds which inhibit iNOS-dependent NO
synthesis without cytotoxic potential.
Nitrite assay (Griess assay)
Experimental
Materials
Cells were seeded in 96-well plates (8Â104/200 mL/well),
cultured for two days, and then incubated with or
without lipopolysaccharide (LPS) in the absence or
presence of the test compounds for 20 h. As a parameter
of NO synthesis nitrite concentration was assessed in
the supernatant of RAW 264.7 macrophages by the
Griess reaction.21 Brie¯y, 100 mL of cell culture super-
natant was removed and combined with (i) 90 mL 1%
sulfanilamide in 5% H3PO4 and (ii) 90 mL 0.1% N-(1-
naphthyl)ethylenediamine dihydrochloride in H2O in a
96-well plate, followed by spectrophotometic measure-
ment at 550 nm (reference wavelength 620 nm) using a
SPECTRA microplate reader (SLT-Labinstruments).
Nitrite concentrations in the supernatants were deter-
mined by comparison with a sodium nitrite standard
curve. Experiments were performed at least three times
and performed in triplicate.
Helenalin and psilotropin were isolated from Psilostrophe
cooperi (A. Gray) Green, ambrosin, hymenin, and bipin-
natin from Hymenoclea salsola Torr. & A. Gray, and tar-
axinic acid-b-d-glucopyranosylester from Taraxacaum
linearisquameum Soest; alantolactone, gaillardin, glauco-
lide A, psilostachyin, and xanthatin were a kind gift from
Dr. E. Rodriguez (Cornell University, Ithaca, NY).
Hydrogenation of alantolactone
Hydrogenation was performed in a Parr hydrogenation
apparatus at room temperature: 10 mg of alantolactone
were dissolved in 15 mL ethyl acetate, 10 mg of Pd on
activated Charcoal (10% Pd, Fluka) were added, and
the mixture was shaken for 18 h under a pressure of 50
psi (=3.5 bar). After ®ltration over Celite the solvent
was removed and the residue was puri®ed by column
chromatography on silica gel using hexane and a mixture
of hexane:ethyl acetate (50:1; 25:1; 10:1; 1:1) as mobile
phases. The obtained fractions were monitored by TLC
(silica gel F254, 0.25 mm (Merck); solvent mixtures:
hexane and hexane:ethyl acetate (10:1); spray reagents:
vanillin (1%) and H2SO4 (10%) in EtOH) and HPLC
[stationary phase: Zorbax Rx-C18 column (4.6 mmÂ25
cm), particle size: 5 mm (Rockland Technology); gra-
dient system: 30±85% MeCN in 20 min; UV detection:
205 nm). The main fraction (6 mg) consisted of 3,5,8a-
trimethyl-3a,5,6,7,8,8a,9,9a-octahydro-3H-naphtho[2,3-
b]furan-2-one and of 3,5,8a-trimethyl-decahydro-3H-
naphtho[2,3-b]furan-2-one (=tetrahydroalanto-lactone)
in a ratio of 45:55. The constitution was established by
1D and 2D NMR experiments and by MS. The NMR
data of both compounds were identical with those pub-
lished earlier.19,20
Cell viability (MTT assay)
Cell respiration, an indicator of cell viability, was
determined by the mitochondrial-dependent reduction
of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium
bromide (MTT) to formazan.22 After removal of super-
natants for nitrite determination cells were incubated at
37 ꢀC with MTT (0.5 mg/mL) for 45 min The medium
was aspirated, and cells were solubilized in DMSO (250
mL) for at least 3 h in the dark. The extent of reduction
of MTT was quanti®ed by OD measurement (550 nm).
NF-ꢀB binding activity (electrophoretic mobility shift
assay)
Cells were seeded in 6-well plates (4Â105/mL), cultured
for two days, and then incubated with or without lipo-
polysaccharide (LPS) in the absence or presence of the
test compounds. Test compounds were pre-incubated 1

V. M. Dirsch et al. / Bioorg. Med. Chem. 8 (2000) 2747±2753
2753
h before LPS stimulation. In order to measure NF-kB
Acknowledgements
binding activity cells were activated with LPS for 1 h.
Experiments set up in parallel to measure nitrite accu-
mulation were performed using an incubation time of
20 h.
We would like to thank U. Ruberg and Ch. Gerhauser
for their excellent technical assistance.
References
Nuclear extracts were prepared as described pre-
viously.23 Brie¯y, cells were washed with PBS, sus-
pended in 400 mL ice cold hypotonic bu€er A (10 mM
HEPES; pH 7.9; 10 mM KCl; 0.1 mM EDTA; 0.1 mM
EGTA; 1 mM DDT; 0.5 mM PMSF), and kept on ice
for 15 min; 25 mL Nonidet NP-40 (10%) was added, the
tubes immediately vortexed for 10 s, and the homo-
genate centrifuged (30 s, 12,000Âg). The pellet was sus-
pended in 50 mL hypertonic bu€er B (20 mM HEPES,
pH 7.9; 0.4 M NaCl; 1 mM EDTA; 1 mM EGTA; 1
mM DTT; 1 mM PMSF) and vigorously rocked for 15
min (4 ꢀC). The extract was centrifuged (5 min,
12,000Âg) and the supernatant (nuclear protein) imme-
diately frozen at 70 ꢀC. Protein concentration was
determined by the Lowry method.24 Nuclear protein
was used for electrophoretic mobility shift assay (EMSA)
as described previously.25 Brie¯y, a double-stranded oli-
gonucleotide containing the most common NF-kB con-
sensus0sequence (22mer, Promega, Mannheim, Germany)
was 5 -end-labeled with [g-32P]-ATP (3000 Ci/mmol;
Amersham, Braunschweig, Germany) using the T4
polynucleotide kinase (Promega, Mannheim, Germany).
Binding reactions were performed incubating 50,000±
200,000 cpm radio-labeled oligonucleotide with nuclear
protein extracts (10 mg protein) at a ®nal reaction
volume of 15 mL (10 mM Tris±HCl, pH 7.5; 1 mM
MgCl2; 50 mM NaCl; 0.5 mM DTT; 4% glycerol, 0.5
mM EDTA; 2 mg poly dI-dC (Promega, Mannheim,
Germany)) for 30 min at room temperature. Nucleo-
protein±oligonucleotide complexes were separated by
1. Spring, O.; Kupka, J.; Maier, B.; Hager, A. Z. Naturforsch.
C 1982, 37, 1087.
2. Chen, H.-C.; Chou, C.-K.; Lee, S.-D.; Wang, J.-C.; Yeh,
S.-F. Antiviral Res. 1995, 27, 99.
3. Beekman, A. C.; Wierenga, P. K.; Woerdenbag, H. J.; Van
Uden, W.; Pras, N.; Konings, A. W. T.; El Feraly, F. S.;
Galal, A. M.; Wikstrom, H. V. Planta Med. 1998, 64, 615.
4. Schinella, G. R.; Giner, R. M.; Del-Carmen-Recio, M.;
Mordujovich-de-Buschiazzo, P.; Rios, J.-L.; Manez, S. J.
Pharm. Pharmacol. 1998, 50, 1069.
5. Lyss, G.; Schmidt, T. J.; Merfort, I.; Pahl, H. L. Biol.
Chem. 1997, 378, 951.
6. Lyss, G.; Knorre, A.; Schmidt, T. J.; Pahl, H. L.; Merfort,
I. J. Biol. Chem. 1998, 273, 33508.
7. Bork, P. M.; Schmitz, M. L.; Kuhnt, M.; Escher, C.;
Heinrich, M. FEBS Lett. 1997, 402, 85.
8. Hehner, S. P.; Heinrich, M.; Bork, P. M.; Vogt, M.; Ratter,
F.; Lehmann, V.; Schulze-Ostho€, K.; Droge, W.; Schmitz, M.
L. J. Biol. Chem. 1998, 273, 1288.
9. Wong, H. R.; Menendez, I. Y. Biochem. Biophys. Res.
Comm. 1999, 262, 375.
10. Ryu, J. H.; Lee, H. J.; Jeong, Y. S.; Ryu, S. Y.; Han, Y.
N. Arch. Pharm. Res. 1998, 21, 481.
11. Lee, H. J.; Kim, N. Y.; Jang, M. K.; Son, H. J.; Kim, K. M.;
Sohn, D. H.; Lee, S. H.; Ryu, J. H. Planta Med. 1999, 65, 104.
12. Forstermann, U.; Kleinert, H. Naunyn-Schmiedeberg's
Arch. Pharmacol. 1995, 352, 351.
13. Clancy, R. M.; Amin, A. R.; Abramson, S. B. Arthritis
Rheum. 1998, 41, 1141.
14. Kupchan, S. M.; Eakin, M. A.; Thomas, A. M. J. Med.
Chem. 1971, 14, 1147.
15. Kupchan, S. M.; Fessler, D. C.; Eakin, M. A.; Giacobbe,
T. J. Science 1970, 168, 376.
electrophoresis on
a 4.5% non-denaturing poly-
16. Xie, Q.-W.; Kashiwabara, Y.; Nathan, C. J. Biol. Chem.
1994, 269, 4705.
17. Zidorn, C.; Dirsch, V. M.; Rungeler, P.; Sosa, S.; Della
Loggia, R.; Merfort, I.; Pahl, H. L.; Vollmar, A. M.; Stupp-
ner, H. Planta Med. 1999, 65, 704.
18. Togashi, H.; Sasaki, M.; Frohman, E.; Taira, E.; Ratan,
R. R.; Dawson, T. M.; Dawson, V. L. Proc. Natl. Acad. Sci.
USA 1997, 94, 2676.
19. Schae€er, M.; Stampf, J.-L.; Benezra, C. J. Org. Chem.
1989, 54, 6106.
acrylamide gel (100 V) and the gel exposed to an X-ray
®lm overnight ( 70 ꢀC, intensifying screen). Speci®city
of the DNA±protein complex was con®rmed by com-
petition with a 100-fold excess of unlabeled NF-kB
(competitor) and AP-2 (non-competitor) oligonucleo-
tides (Promega, Mannheim, Germany), respectively. At
least three independent experiments were performed.
Quanti®cation was performed by densitometry (EASY
plus system, Herolab, Wiesloch, Germany).
20. Nagaki, M. Phytochemistry 1984, 23, 462.
21. Green, L. C.; Wagner, D. A.; Glogowski, J.; Skipper, P.
L.; Wishnok, J. S.; Tannenbaum, S. R. Anal. Biochem. 1982,
126, 131.
22. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
23. Schreiber, E.; Matthias, P.; Muller, M. M.; Scha€ner, W.
Nucleic Acids Res. 1989, 17, 6419.
24. Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R.
J. J. Biol. Chem. 1951, 193, 265.
25. Boese, M.; Busse, R.; Mulsch, A.; Schini-Kerth, V. Br. J.
Pharmacol. 1996, 119, 707.
Statistical analysis
Nitrite determination and cell viability were performed
in triplicate. All experiments including gel-shift assays
were repeated at least two times. Results are expressed
as mean valueÆSEM. Statistical comparisons were
made by ANOVA followed by a Dunnett multiple
comparisons test. P values <0.05 were considered sig-
ni®cant.