Sesquiterpene lactone trilobolide activates production of interferon-γ and nitric oxide

Fitoterapia 81 (2010) 1213–1219

Contents lists available at ScienceDirect

Fitoterapia j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / f i t o t e

Sesquiterpene lactone trilobolide activates production of interferon-γ and nitric oxide Eva Kmoníčková a,b,⁎, Juraj Harmatha c, Karel Vokáč c, Petra Kostecká a,d, Hassan Farghali e, Zdeněk Zídek a a b c d e

Institute of Experimental Medicine, Department of Pharmacology, Academy of Sciences of the Czech Republic, v.v.i., Prague, Czech Republic Institute of Pharmacology and Toxicology, Faculty of Medicine in Pilsen, Charles University, Prague, Czech Republic Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, v.v.i., Prague, Czech Republic 1st Faculty of Medicine, Charles University, Prague, Czech Republic Institute of Pharmacology, 1st Faculty of Medicine, Charles University, Prague, Czech Republic

a r t i c l e

i n f o

Article history: Received 13 April 2010 Accepted in revised form 10 August 2010 Available online 19 August 2010

Keywords: Trilobolide Sesquiterpene lactones SERCA inhibitor Interferon-γ Nitric oxide Laser trilobum

a b s t r a c t Trilobolide (TB), a sesquiterpene lactone isolated from Laser trilobum is an inhibitor of sarco/ endoplasmic reticulum Ca2+-ATPase (SERCA). We have found that upon the in vitro exposure to TB, rodent peritoneal cells and human peripheral blood mononuclear cells secrete high amounts of IFN-γ. The effect is associated with the stimulation of high output NO biosynthesis in rat cells. The stimulatory potential of TB depends on the activation of MAP kinases p38 and ERK1/2, and transcription factor NF-κB. BAPTA-AM, a chelator of the intracellular calcium, remained without any effect on the secretion of IFN-γ triggered by TB. These results demonstrate that TB is a potent immunostimulatory agent. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Sesquiterpene lactones (SLs) are diverse plant secondary metabolites, widely distributed within the family of Asteraceae (Compositae) and Apiaceae (Umbelliferae) [1–3]. They have been receiving an ever increasing attention for prospective anti-inflammatory [4], anti-cancer [5,6], and anti-infectious properties [7–10]. The biological activities of lactones including sesquiterpenoids are largely due to their inhibitory effects on the activity of a plethora of enzymes of prokaryotic organisms and eukaryotic cells [11]. SLs have proved to be potent inhibitors of distinct transcription factors [12,13]. The best recognized interference of SLs with the trans-activating pathways is the

⁎ Corresponding author. Institute of Experimental Medicine, Department of Pharmacology, Academy of Sciences of the Czech Republic, v.v.i., Vídeňská 1083, 142 20, Prague, Czech Republic. Tel./fax: +420 24106 2720. E-mail address: [email protected] (E. Kmoníčková). 0367-326X/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.fitote.2010.08.005

inhibition of multiple factors within the nuclear factor-κB (NF-κB) signaling system [14,15]. These effects are presumed to be common molecular mechanisms determining the prevailing immunosuppressive activity of SLs. The only exception of this rule is the action of thapsigargin (TG), a guaianolide type SL with characteristic lactone-diol moiety isolated from the Mediterranean plant Thapsia garganica L. [16]. It has been shown to stimulate production of interleukin-8 (IL-8) [17] and up-regulate the lipopolysaccharide (LPS)-induced production of pro-inflammatory cytokine IL-6 and tumor necrosis factor-α (TNF-α) [18]. We have found recently that TG is a potent inducer of IFN-γ secretion and activator of nitric oxide (NO) biosynthesis [19]. The aim of the present experiments is to investigate whether this unique property is shared by other structurally related thapsigargin-like SL trilobolide (TB) isolated from Laser trilobum (L.) Borkh. [2,20]. The prominent property of both TG and TB is the inhibition of mammalian sarco/ endoplasmic reticulum Ca2+-ATPase (SERCA) [21]. The

1214

E. Kmoníčková et al. / Fitoterapia 81 (2010) 1213–1219

inhibition of SERCA pumps leads to a release of Ca2+ from intracellular ER stores and a concomitant influx of Ca2+ from extracellular space, resulting in a prolonged elevation of cytosolic Ca2+ concentration. TG is thus widely used experimentally as a tool to elucidate the role of calcium in various biological responses [22]. The present data provide evidence showing that similar to TG, TB is a remarkable stimulator of IFN-γ and high output NO production. The immunostimulatory activity seems to be independent of alterations in intracellular calcium.

2. Materials and methods 2.1. Sesquiterpene lactones Preparation of trilobolide (TB) was based on the previously reported procedures [2] and [23]. The method provides multi-gram preparations as achieved in the protocol reported for thapsigargin [24]. Thapsigargin (TG), used for comparative purposes, was purchased from Sigma. The chemical structures are shown in Fig. 1. They were prepared as a 100 mM stock solution in DMSO. The highest final DMSO concentration in culture wells was 0.02% and was found not to interfere with the assays. The samples were sterile filtered using non-pyrogenic 0.22 μm filters (Costar). The chromogenic Limulus Amoebocyte Lysate assay (KineticQCL; Cambrex Bio Science, Walkersville, MD) was used to check for possible contamination with LPS. The amount of LPS in culture wells containing the highest, i.e. 10 μM concentration of SLs, was b2 pg/mL, the amount that is ineffective to activate production of NO and secretion of cytokines [25].

O O R1

O

H OR2

O

OH OH O O R1

R2

Trilobolide

H

(S)-2-methylbutyryl

Thapsigargin

O-octanoyl

butyryl

Fig. 1. Chemical structure of trilobolide and thapsigargin.

2.2. Other reagents The following selective inhibitors of MAPKs (mitogenactivated protein kinases), obtained from Tocris (Ellisville, MO), were used: a) p38 mitogen-activated protein kinase (p38) inhibitor SB 202190 (4-[4-(4-fluorophenyl)-5-(4-pyridinyl)-1H-imidazol-2-yl]phenol), b) extracellular signalregulated kinases 1/2 (ERK1/2) inhibitor PD 98059 (2-(2amino-3-methoxyphenyl)-4H-1-benzopyran-4-one), and c) cJun N-terminal kinase (JNK) inhibitor SP 600125, i.e. anthra [1-9-cd]pyrazol-6(2H)-one. The working RPMI-1640 solutions of the inhibitors were prepared from the 10 mM stock DMSO solutions. The final 10 μM concentrations were not cytotoxic (data not shown). The inhibitor of transcriptional factor NF-κB, pyrrolidine dithiocarbamate (PDTC; Sigma), was prepared as a 5 mM stock solution in PBS. In some experiments, the culture medium was supplemented with arginine and sepiapterin. Arginine, a substrate for NO synthesis, was from Sigma. Sepiapterin (2-amino-4hydroxy-6-lactoyl-7,8-dihydropterin), a pteridine substituting via a salvage pathway the tetrahydropterin requirements of inducible NO synthase (iNOS) to synthesize NO [26], was obtained from Sigma. The 100 mM stock solution was prepared in DMSO. In order to deplete intracellular calcium, the calcium chelator BAPTA-AM (1,2-bis(2-aminophenoxy)-ethane-N,N, N′,N′-tetraacetic acid (acetoxymethyl) ester) (Sigma), was used. Its possible toxicity under the conditions of present experiments was determined (see Results). Biological effectiveness of 10 μM BAPTA-AM has been demonstrated in many studies [27–29]. 2.3. Biological material, in vitro cell cultures Immunobiological activities have been determined under the conditions in vitro using rat and mouse resident peritoneal cells (PEC), and human peripheral blood mononuclear cells (PBMC). All protocols were approved by the Institute Ethics Committee (no. 13/2006). Female rats of the inbred strain LEWIS, 165–180 g of weight, and female mice of the inbred strain C57BL/6, 8–11 weeks old, were purchased from Charles River Deutschland (Sulzfeld, Germany). They were kept in transparent plastic cages and maintained in an Independent Environmental Air Flow Animal Cabinet (ESI Flufrance, Wissous, France). Lighting was set on 06 to 18 h, temperature at 22 °C. Animals, killed by cervical dislocation, were i.p. injected with 8 mL (mice) or 16 mL (rats) of sterile saline. Pooled PEC collected from 4 to 8 mice, and PEC from individual rats were washed, resuspended in culture medium and seeded into the 96-well round-bottom culture microplates (Costar, Cambridge, MA) in final 100-μl volumes, 2 × 106/mL. The human PBMCs were separated by Ficoll-Paque gradient centrifugation (GE Healthcare Bio-Sciences, AB, Uppsala, Sweden) according to the manufacturer instructions. The buffy coats acquired from healthy donors were provided by the Institute of Hematology and Blood Transfusion, Prague. The human PBMCs were cultured at a final density of 1.0 × 106/mL. The cell cultures were maintained at 37 °C, 5% CO2 in humidified Heraeus incubator. The complete RPMI-1640

E. Kmoníčková et al. / Fitoterapia 81 (2010) 1213–1219

culture medium contained 10% heat-inactivated foetal bovine serum, 2 mM L-glutamine, 50 μg/mL gentamicin, and 5 × 10−5 M 2-mercaptoethanol (all Sigma-Aldrich, Prague, CR).

1215

3. Results 3.1. Secretion of IFN-γ

The concentration of nitrites was detected in individual, cell-free samples (50 μl) incubated 5 min at an ambient temperature with an aliquot of a Griess reagent (1% sulphanilamide/0.1% naphtylendiamine/2.5% H3PO4). The absorbance at 540 nm was recorded using a microplate spectrophotometer (Tecan, Austria). A nitrite calibration curve was used to convert absorbance to μM nitrite. Concentration of IFN-γ (pg/mL) in supernatants of cells was determined by ELISA kit, following the manufacturer instructions (R&D Systems, Abingdon, UK).

TB, applied within the range of 0.04–10 μM concentrations, activated the in vitro secretion of IFN-γ in all rat, mouse and human cells (Fig. 2A). The most prominent effect was observed in rat PEC in which as low concentration as 40 nM was effective. The significant response in human PBMC was produced with the 1 μM dose of TB. Statistically significant though low amount of IFN-γ (b100 pg/mL) also was observed in mouse PEC. The highest accumulation of IFN-γ in supernatants of mouse PEC and human PBMC was achieved with the dose of 10 μM TB. The stimulation of IFN-γ secretion plateaued with approximately 1 μM TB concentration in rat PEC. The onset of significantly enhanced IFN-γ production induced by 5 μM TB was observed at the intervals of 5 h and 16 h in rat PEC and human PBMC, respectively (Fig. 2B).

2.5. Cell viability assay

3.2. Production of NO

Viability of cells was determined using a colorimetric assay based on the cleavage of the tetrazolium salt WST-1 by mitochondrial dehydrogenases in viable cells (Roche Diagnostics, Mannheim, Germany). The cells (1 × 106/mL) were cultured in quadruplicate under the conditions described above. After 5 or 24 h of culture, the WST-1 was added and the cells were kept in the Heraeus incubator at 37 °C for additional 3 h. Optical density at 450–690 nm was evaluated.

Production of NO was evaluated after the 24-h culture of cells in the presence of increasing concentrations of TB (0.04– 5 μM) (Fig. 3). The stimulatory activity was only marginal, though statistically significant, in mouse PEC. No NO production was observed in human PBMC (data not shown). Starting with 0.1 μM TB, rat PEC produced significantly enhanced amounts of NO (21.7 ± 3.3 μM vs. 9.2 ± 1.8 μM in controls; P b 0.001). It was increasing towards the 0.4 μM TB concentration. At this point, the NO levels were approximately 4-fold higher than the control values (35.7± 3.1 μM vs. 9.2± 1.8 μM in controls; P b 0.001). The increasing concentrations of TB (1 and 5 μM) were less effective to stimulate production of NO than was the 0.4 μM concentration. The effects of higher i.e. 1 μM and 5 μM TB concentrations resembled those produced by 0.2 μM and 0.1 μM TB doses, respectively (Fig. 3). The bell-shaped curve characterizing the dependence of NO production on TB

2.4. NO and IFN-γ assay

2.6. Statistical analysis Analysis of variance (ANOVA) with subsequent Newman– Keuls or Bonferroni's multiple comparison tests and graphical presentation of data were done using the Prism program (GraphPad Software, San Diego, CA).

(A)

2000

Rat PEC: +Trilobolide 5.0 μM Human PBMC: +Trilobolide 5 µM 3000

Mouse PEC: Untreated Human PBMC: Untreated Rat PEC: Untreated

***

1000 ***

**

Rat PEC: Untreated Human PBMC: Untreated

2000

***

1000

*

100

Trilobolide 5 µM

4000

Rat PEC: +Trilobolide Human PBMC: +Trilobolide Mouse PEC: +Trilobolide

IFN-γ (pg/ml)

IFN-γ (pg/ml)

3000

(B)

24-h culture

4000

***

50 0

Controls10-2

10-1

*

*** 100

Trilobolide (µM)

101

0 0

4

8

12

16

20

24

Length of culture (h)

Fig. 2. In vitro activation of interferon-gamma (IFN-γ) secretion by trilobolide. Rat and mouse peritoneal cells (PEC) and human peripheral blood mononuclear cells (PBMC) were cultured in the presence of increasing concentration of trilobolide for 24 h (A) or for indicated time intervals in the presence of 5 μM trilobolide (B). The supernatant levels of IFN-γ are both dose- and time-dependent. Data are means ± S.E.M. obtained from 4 rats, 4 mice and 4 donors of human cells. Statistical significance was determined using Newman–Keuls test (*P b 0.05; ***P b 0.001).

1216

E. Kmoníčková et al. / Fitoterapia 81 (2010) 1213–1219

40

The IFN-γ secretion induced by 0.4 μM TB (2770±199 pg/mL) was substantially reduced by the p38 MAPK inhibitor SB202190 (by 92%; 226.0±42.0 pg/mL), and also by the ERK1/2 MAPK inhibitor PD98059 (by 77%; 645.5 ±53.5 pg/mL). The inhibitor of JNK MAPK was ineffective to suppress the IFN-γ secretion. Production of both IFN-γ and NO induced in rat PEC by TB was dose-dependently inhibited by PDTC, the inhibitor of transcription factor NF-κB (Fig. 5B). The IC50 estimates (±95% limits of confidence) for the inhibition of IFN-γ and NO production were almost identical, reaching the values of 3.57 μM (3.18–4.01 μM) and 3.50 μM (2.84–4.31 μM), respectively.

Rat PEC: Untreated Mouse PEC: Untreated *** Rat PEC: +Trilobolide Mouse PEC: +Trilobolide ***

Nitrite (µM)

30

***

20 *

*

10 ** 0

Controls 10-2

** 100

10-1

101

3.5. Cytotoxicity of BAPTA-AM

Trilobolide (μM) Fig. 3. Trilobolide-triggered in vitro production of nitric oxide by rat and mouse peritoneal cells (PEC; 2 × 106/mL). Concentration of nitrites was determined after the 24-h cell culture. Each point is the mean ± S.E.M. The data are representative of five independent experiments (n = 3–4, each). Statistical significance was determined using Newman–Keuls multiple comparison test (*P b 0.05, **P b 0.01, ***P b 0.001).

concentration remained unchanged after the supplementation of rat PEC cultures with L-arginine (Fig. 4A) or sepiapterin (Fig. 4B).

We have observed that in dependence on the dose and time interval, BAPTA-AM may be cytotoxic to rat PEC. The cytocidal effect was more pronounced after 24 h than after 6 h of culture (Fig. 6). While 10 μM BAPTA-AM had only a marginal effect at the interval of 6 h, it decreased the cell viability at the interval of 24 h (by 37%: OD450–690 nm for controls was 1.811 ± 0.063, and that for TB was 0.670 ± 0.028; P b 0.001). The 20 μM concentration of BAPTA-AM decreased the cell viability at both 6-h (by 33%: OD450–690 nm for controls was 1.511±0.055, and that for TB was 0.559±0.028; P b 0.001) and 24-h (by 75%: OD450–690 nm for controls was 1.811±0.063, and that for TB was 0.241±0.14; Pb 0.001) intervals.

3.3. Viability of cells 3.6. Effects of BAPTA-AM on IFN-γ and NO production Viability of rat PEC was evaluated after their 24-h exposure to TB. It was not significantly influenced up to the concentration of 5 μM TB. Significantly decreased viability by 23 ± 0.95% of the rat PEC (OD450–690 nm for controls was 1.7650 ± 0.082, and that for TB was 1.362 ± 0.017; P b 0.01) was observed with 10 μM TB. No suppression of viability of human PBMC was found after the dose of 10 μM TB (data not shown).

Highly enhanced levels of IFN-γ that were produced by rat PEC stimulated for 6 h with TB or TG were not suppressed by the Ca2+ chelating agent BAPTA-AM applied at doses up to 10 μM. The same doses of BAPTA-AM inhibited production of NO determined at the interval of 24 h. The relatively high (toxic) dose of BAPTA-AM (20 μM) completely abrogated NO production and reduced production of IFN-γ (Fig. 7).

3.4. Involvement of MAP kinases and NF-κB 4. Discussion The participation of major MAPKs in the expression of immunostimulatory effects of TB was investigated using the specific pharmacological inhibitors of MAPK activity (Fig. 5A).

Many SLs have proved to possess immunosuppressive activities. Artemisinin [30], costunolide [31,32], helenalin [13],

(B)

(A) 40

40

TB alone TB+Arginine 1 mM TB+Arginine 2 mM

Untreated Arginine 1 mM Arginine 2 mM

30

Nitrite (µM)

Nitrite (µM)

30

20

10

0

Untreated Sepiapterin 100 µM Sepiapterin 250 µM

TB alone TB+Sepiapterin 100 µM TB+Sepiapterin 250 µM

20

10

Controls10-1

100

Trilobolide (TB), µM

101

0

Controls10-1

100

101

Trilobolide (TB), µM

Fig. 4. Biosynthesis of trilobolide (TB)-triggered nitric oxide (NO) in the presence of L-arginine (A) and sepiapterin (B). L-arginine and sepiapterin were added 10 min before TB. Production of NO was determined after the 24-h culture of peritoneal cells (2 × 106/mL) pooled from three rats (each experiment). Data are means ± S.E.M. representing one of two identical experiments.

E. Kmoníčková et al. / Fitoterapia 81 (2010) 1213–1219

(A)

1217

(B)

3000 IFN-γ: Untreated cells Secretion of IFN-γ

***

n.s.

NO: Untreated cells Production of NO 20

2000 Without PDTC

SP 600125 (JNK)

***

Nitrite (µM)

PD 98059 (ERK1/2)

0

SB 202190 (p38)

Trilobolide 0.4 µM alone

1000

Untreated cells

2000

3000

IFN-γ (pg/ml)

IFN-γ (pg/ml)

30

10

0

1000

0 0

Trilobolide 0.4 µM + MAPK inhibitiors (5 µM)

5

10

15

20

25

Trilobolide 0.4 µM + PDTC (µM)

Fig. 5. Dependence of immunostimulatory effects of trilobolide on mitogen-activated protein kinases (MAPKs) (A) and nuclear factor-κB (NF-κB) (B) activity. Rat peritoneal cells were obtained from four rats and cultured (2 × 106/mL) in the presence of trilobolide (0.4 μM). The MAPK inhibitors and pyrrolidine dithiocarbamate (PDTC) were applied 30 min and 15 min before trilobolide, respectively. Production of nitric oxide (NO) and interferon-γ (IFN-γ) was determined 24 h afterwards. The bars are means ± S.E.M. The data are representative of two identical experiments. Statistical significance was determined using Newman–Keuls test (n.s., not significant, ***P b 0.001).

factor) may be relevant to the antiangiogenic and anti-cancer properties of SLs. Also the biosynthesis of NO which is tightly regulated by cytokines and crucially depends on the activity of NF-κB [43] is inhibited by SLs of distinct skeleton structures such as costunolide [44], helenalin [45], chamissonolide [13], isohelenin [46], parthenolide [46], eupatolide, sesquiterpene dimers inulanolides A–D [47], eremanthine, magnolialide, santamarine, spirafolide, zaluzanin C [48], and ergolide [49]. Potent inhibitors of NO production are artemisinin and its derivatives [50]. Arthemether diminishes NO formation in arthritic rats [51]. The present findings are at striking variance with these effects. We have found that TB activates production of IFN-γ in both animal and human cells. Since IFN-γ is the major stimulus

100 IFN-γ production:

60 40 Control 20

6-h culture 24-h culture

0 1

10

(in % of controls)

100

80

Production of NO and IFN-γ

OD 450-690 nM (in % of lortnoc)

parthenolide [33], and budlein A [34] suppress the expression of immune-activated and/or disease-associated pro-inflammatory cytokines IL-1β, IL-6, and TNF-α. The secretion of Th1 cytokines IFN-γ and IL-2 is inhibited by artemisinin [30,35], helenalin [36], and parthenolide [37]. Parthenolide suppresses production of IL-12 [38] and the Th2 cytokine IL-4 [37]. The secretion of chemoattractant cytokines (chemokines) is suppressed by SLs as well. The IL-8 and MCP-1 (monocyte chemoattractant protein-1) are inhibited by parthenolide [39,40], MIP-1α (macrophage inflammatory protein-1α) by artemisinin [41], and secretion of KC (keratinocyte-derived chemokine) by budlein A [34]. The capability of SLs such as artemisinin [42] and parthenolide [39] to inhibit expression of the major angiogenic factor VEGF (vascular endothelial growth

TB 0.4 μM TG 0.4 μM

80 60 40 NO production: TB 0.4 μM

20

100

BAPTA-AM (µM) Fig. 6. Cytotoxicity of BAPTA-AM (1,2-bis(2-aminophenoxy)-ethane-N,N,N′,N′tetraacetic acid (acetoxymethyl) ester). Changes in viability of rat peritoneal cells (1 × 106/mL) were evaluated 6 h and 24 h after addition of BAPTA-AM. The optical density characterizing the cleavage of the tetrazolium salt WST-1 was determined after the successive 3-h supplementation of cell cultures with WST1 (see Materials and methods). The points are means± S.E.M. showing the effects expressed in percent of untreated controls. Cytotoxic effects of BAPTAAM were more pronounced at the interval of 24-h (IC50 = 12.10 μM; 95% limits of confidence = 10.56–13.86 μM) than at the interval of 6-h culture (IC50 = 31.42 μM; 95% limits of confidence = 28.41–34.75 μM).

TG 0.4 μM

0 0

5

10

15

20

BAPTA-AM (μM) Fig. 7. Effects of calcium chelator BAPTA-AM on nitric oxide (NO) and interferongamma (IFN-γ) production induced by trilobolide (TB) and thapsigargin (TG). The effects are expressed in percent of the effects of TB and TG alone. Rat peritoneal cells (2 ×106/mL) were cultured and NO was determined at the interval of 24 h of culture (100%=31.0±0.25 μM, and 27.1± 0.13 μM, respectively). Secretion of IFN-γ was assayed at the interval of 6 h following addition of TB and TG (100%= 2708 ±140 pg/mL, and 2801±16 pg/mL, respectively). The points are means±S.E.M. representing one of two similar experiments.

1218

E. Kmoníčková et al. / Fitoterapia 81 (2010) 1213–1219

which directly activates biosynthesis of NO from L-arginine [52], it is only conceivable that administration of TB also leads to the activation of NO production. Virtually the same immunobiological activities are possessed by TG [19]. The similarities in the action of both TB and TG encompass a plethora of pharmacodynamic patterns. While the IFN-γ production increases steadily with the increasing concentration of both TB and TG, the production of NO follows the bell-shaped dose–response curve. The drop in the NO biosynthesis observed in higher concentrations of these agents is not due to their cytotoxicity. Neither it is due to possible interference with the arginine transport and tetrahydrobiopterine formation, key prerequisites for the effective NO biosynthesis. Furthermore, the previous experiments [19] have shown that this phenomenon does not result from changes in the activity of arginase, an enzyme which can competitively metabolize L-arginine to urea and ornithine [53]. The mechanistic background for the bell-shaped curve, typical for the dependence of NO production on a concentration of TG and TB, thus remains to be elucidated. Interestingly, similar biphasic mode of action has been observed in the regulation of reactive oxygen species (ROS) and caspase-3 activity by TG [54]. The up-regulatory effects of both TG and TB on IFN-γ and NO production are only marginally expressed in peritoneal cells of mice. Plausible, though speculative explanation might be based on the findings showing remarkable differences in the expression of various SERCA3a isoforms in mice, rats and humans that may be differentially inhibited by the compounds [55]. In contrast to the prominent stimulatory effects on IFN-γ production, neither TB nor TG activate NO production in human PBMCs. This failure is however not surprising in view of the notoriously recognized inability of human cells to produce NO under conditions in vitro [56]. The NO production and IFN-γ secretion triggered by both TG [19] and TB can be inhibited by PDTC, an inhibitor of NFκB. The data are in good keeping with the findings showing that TG on its own is an inducer of NF-κB in various cell types such as HeLa cells [57], Jurkat T cells [58], and rat peritoneal macrophages [59]. The present data suggest that TB also activates p38 and ERK1/2 MAP kinases. As far as the intracellular free Ca2+ levels modulate cellular signaling and gene expression [60], it would be tempting to consider this effect as a plausible mechanism explaining the immunostimulatory activity of TB and TG, modulators of intracellular calcium. However, we have found that the intracellular Ca2+ chelating agent BAPTA-AM applied at the non-toxic concentrations (up to 10 μM) is ineffective to suppress the TB-induced IFN-γ secretion, assayed at the early interval of 6 h of the cell culture. Suppression of IFN-γ production only was observed with the higher (20 μM), already cytotoxic concentration of BAPTA-AM. The BAPTAAM cytotoxicity was greatly enhanced at the interval of 24 h. This may at least partially explain the inhibitory effects of BAPTA-AM on production of NO by cells exposed to BAPTAAM for 24 h. The NO biosynthesis as well as the expression of inducible NO synthase (iNOS) mRNA and its maximal enzyme activity have been shown to be regulated by intracellular calcium [61,62]. Therefore, the additional mechanism of the inhibition of NO may be the chelation of Ca2+ by BAPTA-AM.

In conclusion, the sesquiterpene lactone trilobolide (TB) is a potent activator of IFN-γ secretion and high output NO production. The lack of BAPTA-AM to interfere with the secretion of IFN-γ suggests that the immunostimulatory effects of TB are not mediated by mobilization of cytosolic calcium resulting from the inhibition of SERCA. Both IFN-γ and NO play key roles in anti-infectious immune defense mechanisms [63]. The findings encourage further studies to analyze full range of possible immunostimulatory properties of structurally transformed derivatives of TB. For this purpose a series of deacylated and relactonized derivatives were prepared by alkaline hydrolysis of TB [64]. In addition, also the implication of the diol moiety combined with other structure functionality changes was assessed. Our preliminary data show that none of the deacylated derivatives retained the original activities of TG [65]. The significantly changed hydrophobicity of the TB deacyl derivatives, with partially or totally removed side chains, might enforce the inactivation role in the immunomodulatory properties, as did in the SERCA inhibition the TG analogs in a related study [66]. Acknowledgement The study was supported by grant no. 305/07/0061 from the Grant Agency of the Czech Republic. References [1] Fischer NH, Olivier EJ, Fischer HD. Prog Chem Org Nat Prod 1979;38:47. [2] Holub M, Buděšínský M. Phytochemistry 1986;25:2015. [3] Drew DP, Krichau N, Reichwald K, Simonsen HT. Phytochem Rev 2009;8:581. [4] Wang JX, Tang W, Zhou R, Wan J, Shi LP, Zhang Y, et al. Br J Pharmacol 2008;153:1303. [5] Zhang S, Won YK, Ong CN, Shen HM. Curr Med Chem Anticancer Agents 2005;5:239. [6] Christensen SB, Skytte DM, Denmeade SR, Dionne C, Møller JV, Nissen P, et al. Anticancer Agents Med Chem 2009;9:276. [7] Tiuman TS, Ueda-Nakamura T, Garcia Cortez DA, Dias Filho BP, Morgado-Díaz JA, de Souza W, et al. Antimicrob Agents Chemother 2005;49:176. [8] Vu TK, Tuyen NV, Sung TV. Bioorg Med Chem Lett 2005;15:2629. [9] Hwang D-R, Wu Y-S, Chang C-W, Lien T-W, Chen W-C, Tan U-K, et al. Bioorg Med Chem 2006;14:83. [10] Boulanger D, Brouillette E, Jaspar F, Malouin F, Mainil J, Bureau F, et al. Vet Microbiol 2007;119:330. [11] Konaklieva MI, Plotkin BJ. Mini Rev Med Chem 2005;5:73. [12] Dirsch VM, Stuppner H, Ellmerer-Müller EP, Vollmar AM. Bioorg Med Chem 2000;8:2747. [13] Gertsch J, Sticher O, Schmidt T, Heilmann J. Biochem Pharmacol 2003;66:2141. [14] Siedle B, García-Piñeres AJ, Murillo R, Schulte-Mönting J, Castro V, Rüngeler P, et al. J Med Chem 2004;47:6042. [15] Salminen A, Lehtonen M, Suuronen T, Kaarniranta K, Huuskonen J. Cell Mol Life Sci 2008;65:2979. [16] Christensen SB, Andersen A, Smitt UW. Prog Chem Org Nat Prod 1997;71:129. [17] Gewirtz AT, Rao AS, Simon POJ, Merlin D, Carnes D, Madara JL, et al. J Clin Invest 2000;105:79. [18] Chen B-C, Hsieh S-L, Lin W-W. J Leukoc Biol 2001;69:280. [19] Kmoníčková E, Melkusová P, Harmatha J, Vokáč K, Farghali H, Zídek Z. Eur J Pharmacol 2008;588:85. [20] Smítalová Z, Buděšínský M, Šaman D, Holub M. On Terpens 291. Collect Czech Chem Commun 1986;51:1323. [21] Søhoel H, Lund Jensen A-M, Møller JV, Nissen P, Denmeade SR, Isaacs JT, et al. Bioorg Med Chem 2006;14:2810. [22] Treiman M, Caspersen C, Christensen SB. TIPS 1998;19:131. [23] Nalepka JL, Greenfield EM. Biotechniques 2004;37:413. [24] Pagani A, Pollastro F, Spera S, Ballero M, Sterner O, Appendino G. Nat Prod Commun 2007;2:637.

E. Kmoníčková et al. / Fitoterapia 81 (2010) 1213–1219 [25] Zídek Z, Potměšil P, Kmoníčková E, Holý A. Eur J Pharmacol 2003;475: 149. [26] Schoedon G, Schneemann M, Hofer S, Guerrero L, Blau N, Schaffner A. Eur J Biochem 1993;213:833. [27] Gonzáles-Benítez E, Guinzberg R, Díaz-Cruz A, Piña E. Eur J Pharmacol 2002;437:105. [28] Kim J, Sharma RP. Toxicol Sci 2004;81:518. [29] Xu W, Liu L, Charles IG, Moncada S. Nat Cell Biol 2004;6:1129. [30] Wang J, Zhou H, Zheng J, Cheng J, Liu W, Ding G, et al. Antimicrob Agents Chemother 2006;50:2420. [31] Kang JS, Yoon YD, Lee KH, Park SK, Kim HM. Biochem Biophys Res Commun 2004;313:171. [32] Pae HO, Jeong GS, Kim HS, Woo WH, Rhew HY, Kim HS, et al. Inflamm Res 2007;56:520. [33] Hwang D, Fischer NH, Jang BC, Tak H, Kim JK, Lee W. Biochem Biophys Res Commun 1996;226:810. [34] Valério DA, Cunha TM, Arakawa NS, Lemos HP, Da Costa FB, Parada CA, et al. Eur J Pharmacol 2007;562:155. [35] Wang JX, Tang W, Shi LP, Wan J, Zhou R, Ni J, et al. Br J Pharmacol 2007;150:652. [36] Humar M, García-Piñeres AJ, Castro V, Merfort I. Biochem Pharmacol 2003;65:1551. [37] Li-Weber M, Giaisi M, Treiber MK, Krammer PH. Eur J Immunol 2002;32:3587. [38] Kang BY, Chung SW, Kim TS. Immunol Lett 2001;77:159. [39] Oka D, Nishimura K, Shiba M, Nakai Y, Arai Y, Nakayama M, et al. Int J Cancer 2007;120:2576. [40] Bian ZM, Elner SG, Yoshida A, Kunkel SL, Su J, Elner VM. Exp Eye Res 2001;73:111. [41] Zhao M, Xue DB, Zheng B, Zhang WH, Pan SH, Sun B. World J Gastroenterol 2007;13:5612. [42] Wang J, Zhang B, Guo Y, Li G, Xie Q, Zhu B, et al. Pharmacology 2008;82: 148. [43] Xie Q-w, Kashiwabara Y, Nathan C. J Biol Chem 1994;269:4705. [44] Lee HJ, Kim NY, Jang MK, Son HJ, Kim KM, Sohn DH, et al. Planta Med 1999;65:104. [45] Davis RL, Sanchez AC, Lindley DJ, Williams SC, Syapin PJ. Nitric Oxide 2005;12:200.

1219

[46] Wong HR, Menendez IY. Biochem Biophys Res Commun 1999;262:375. [47] Jin HZ, Lee D, Lee JH, Lee K, Hong YS, Choung DH, et al. Planta Med 2006;72:40. [48] Matsuda H, Kagerura T, Toguchida I, Ueda H, Morikawa T, Yoshikawa M. Life Sci 2000;66:2151. [49] Whan Han J, Gon Lee B, Kee Kim Y, Woo Yoon J, Kyoung Jin H, Hong S, et al. Br J Pharmacol 2001;133:503. [50] Aldieri E, Atragene D, Bergandi L, Riganti C, Costamagna C, Bosia A, et al. FEBS Lett 2003;552:141. [51] Cuzzocrea S, Saadat F, Di Paola R, Mirshafiey A. Immunopharmac Immunotoxicol 2005;27:615. [52] Stuehr DJ, Marletta MA. J Immunol 1987;139:518. [53] Sonoki T, Nagasaki A, Gotoh T, Takiguchi M, Takeya M, Matsuzaki H, et al. J Biol Chem 1997;272:3689. [54] Yip KH, Zheng MH, Steer JH, Giardina TM, Han R, Lo SZ, et al. J Bone Miner Res 2005;20:1462. [55] Bobe R, Bredoux R, Corvazier E, Lacabaratz-Porret C, Martin V, Kovács T, et al. Platelets 2005;16:133. [56] Zhang X, Laubach VE, Alley EW, Edwards KA, Sherman PA, Russell SW, et al. J Leukoc Biol 1996;59:575. [57] Pahl HL, Sester M, Burgert HG, Baeuerle PA. J Cell Biol 1996;132:511. [58] Sei Y, Reich H. Immunol Lett 1995;45:750. [59] Tsuyuki K, Ichinowatari G, Tanimoto A, Yamada M, Yaginuma H, Ohuchi K. Biochem Biophys Acta 2002;1583:26. [60] Mellström B, Savignac M, Gomez-Villafuertes R, Naranjo JR. Physiol Rev 2008;88:421. [61] Schmidt HHHW, Warner TD, Nakane M, Förstermann U, Murad F. Mol Pharmacol 1992;41:615. [62] Jordan ML, Rominski B, Jaquins-Gerstl A, Geller D, Hoffman RA. Surgery 1995;118:138. [63] Nathan CF, Hibbs JB. Curr Opin Immunol 1991;3:65. [64] Harmatha J, Vokáč K, Kmonícková E, Zídek Z. Planta Med 2008;74:264. [65] Zídek Z, Harmatha J, Vokáč K, Kmoníčková E. Planta Med 2009;75:905. [66] Winther A-M, Liu H, Sonntag Y, Olesen C, le Maire M, Soehoel H, et al. J Biol Chem 2010 [Epub ahead of print].