Bio-guided identification of an anti-inflammatory guaianolide from Centaurea ainetensis

Abstract

Inflammatory processes are characterized by the stimulation of humoral and cellular mediator systems resulting in increased levels of mediators of inflammation such as interleukins (ILs) and prostaglandins (PGs), which play central roles in regulating the inflammatory response and inflammation-mediated damage. The aim of the study was to develop natural therapeutics for the interventions and/or prevention of inflammatory disorders. A methanol plant extract of Centaurea ainetensis Boiss. (Asteraceae) was tested for its anti-inflammatory potential activity in two cell models of inflammation, which served as the basis for the bio-guided fractionation of the extract and the identification of its active ingredient salograviolide A. The guaianolide salograviolide inhibited endotoxin (ET)-induced IL-6 levels in SCp2 mammary epithelial cells and decreased the levels of IL-1-induced cyclooxygenase enzyme levels in intestinal epithelial cells. The results provide evidence for the anti-inflammatory activities of the sesquiterpene agent salograviolide A and the biochemical basis for further studies on salograviolide A mechanism of action.

Keywords::

• Anti-inflammatory
• Asteraceae
• Centaurea ainetensis
• guaianolides
• intestinal cells
• mammary cells
• salograviolide A

Introduction

Centaurea ainetensis Boiss., known in Lebanon as “Chaouk El Dardar” or “Qanturyun aynata”, belongs to one of the largest genera in the Asteraceae family, and is used in Lebanese folk medicine (Talhouk et al., 2007). It grows in stony, sterile, or bushy places, and the flowering period extends from May to June (Post & Dinsmore, 1932; Talhouk et al. 2008). Many species of the genus Centaurea (C. chilensis Bert. Ex Bull., C. cyanus L., and C. scabiosa L.) have long been used in folk medicine for the treatment of several ailments (Kumarasamy et al., 2003) including rheumatism (Helal et al., 1997; Negrete et al., 1993), malaria, hypertension, diarrhea (Kumarasamy et al., 2002), cough (Flamini et al., 2001), and diabetes (Helal et al., 1997; Kumarasamy et al., 2002).

Extensive chemical investigations of Centaurea species led to the isolation and identification of various types of compounds, including alkaloids (Sarker et al., 2001), lignans (Aslan & Oksuz, 1999; Kumarasamy et al., 2003), acetylenes, flavonoids (Akkal et al., 1997, 1999, 2003; Aslan & Oksuz, 1999; de Almeida et al., 1998; Flamini et al., 2001, 2002), and sesquiterpene lactones (Aslan & Oksuz, 1999; Bruno et al., 2002; Chaves & De Oliviera, 2003; Fortuna et al., 2001, 2002; Helal et al., 1997; Medjroubi et al., 1997, 1998; Oksuz et al., 1994; Skaltsa et al., 2000; Vajs et al., 1999; Youssef, 1998). It has been reported that the most characteristic constituents of the genus Centaurea are sesquiterpene lactones, which are classified into guaianolides, the most abundant in this genus, followed by germacranolides, elemenolides, and finally eudesmanolides (Skaltsa et al., 2000).

Recently, we have reported that crude water extracts from C. ainetensis possessed anti-inflammatory bioactivity that reduced interleukin-6 (IL-6) and gelatinases A and B expression by endotoxin (ET)-treated mammary epithelial cells. The extract also reduced paw edema and thermal hyperalgesia in rats receiving intraplantar injections of ET (Talhouk et al., 2008). This study focuses on the isolation and structure identification of an anti-inflammatory agent from the aerial parts of C. ainetensis following a bioassay-guided fractionation procedure. Detailed description of the fractionation and identification procedures in parallel with the anti-inflammatory assays is presented.

Materials and methods

General experimental procedure

All deuterated and non-deuterated solvents were purchased from Acros Organics, Belgium, and the preparative thin layer chromatography (TLC) plates, silica cartridges, and silica gel were obtained from Alltech Associates, Inc., Pennsylvania, USA. Dulbecco’s modified Eagle’s medium nutrient mixture F12 Ham (DMEM/F12), phosphate-buffered saline (PBS), and heat inactivated fetal bovine serum (FBS), glucose, sodium pyruvate, trypsin, Trypan blue dye, and non-essential amino acids were obtained from Gibco, Paisley, Scotland. Insulin, gentamycin, hydrocortisone, and ovine prolactin were obtained from Sigma, St. Louis, USA; Escherichia coli, strain 055:B5, endotoxin (ET; lipopolysaccharide) was purchased from Difco Laboratories (Detroit, MI, USA). Complete Protease Inhibitor Cocktail was obtained from Roche Diagnostics GmbH. IL-1 was obtained from US Biological, Swampscott, MA, USA. Tris-HCl, sodium dodecyl sulfate (SDS), glycerol, polyacrylamide, bisacrylamide, glycine, Tween 20, polyvinylidene fluoride (PVDF) membranes, and bromophenol blue were obtained from Amersham BioSciences through Numelab (Beirut, Lebanon).

Plant material

The aerial parts of Lebanese C. ainetensis (= C. eryngioides Lam. var. ainetensis Bois.) (Post & Dinsmore, 1932) were collected during the flowering season of the plant in the months of May and June 2002–2006. The plant material was identified by Dr. Stephen Jury at the University of Reading, UK. A voucher specimen has been deposited in the Post Herbarium (BEI), at the American University of Beirut. The aerial parts were dried by leaving the plant sample in the shade for 2 weeks before grinding it into around 10 mm pieces using a blender.

Extraction and purification

The ground plant material (250 g) was soaked in 2500 mL of methanol (MeOH) (1:10 w/v) for 16 h, during which the mixture was incubated on a shaker at 20°C. The extraction scheme was adopted from Harborne (1998) and is shown in Figure 1. The methanol extract was filtered to give a residue labeled as “R-I” and a filtrate labeled as “I” and named as “methanol crude extract”. R-I was then soaked in ethyl acetate (EtOAc) (10:1 w/v) and filtered to give a filtrate, labeled “I.1”. To the crude methanol extract “I”, concentrated sulfuric acid solution was added dropwise untill the pH reached 2. Following acidification, a mixture of chloroform (CHCl₃):water (2:1 v/v) was added. The chloroform phase was collected and labeled as “I.2”. The aqueous layer was then basified to pH 10 by adding concentrated ammonium hydroxide solution dropwise and then resuspended in CHCl₃:MeOH mixture (3:1 v/v). It was later separated into organic and aqueous layers labeled “I.3” and “I.4”, respectively. All these fractions were concentrated in vacuo at 40°C; residues were weighed and then dissolved in a minimal amount of ethanol or dimethylsulfoxide (DMSO) and were assayed for their anti-inflammatory bioactivity against ET-treated SCp2 cells. Only fraction I.2, which exhibited anti-inflammatory activity, was concentrated in vacuo at 40°C, and the obtained residue (0.48 g) was applied to a chromatographic column consisting of 150 g of silica gel (0.035–0. 07 mm and 6 nm pore diameter). A gradient elution was performed using petroleum ether:CHCl₃:EtOAc (2:2:1) (3000 mL), followed by petroleum ether:CHCl₃:EtOAc (1:3:1) (500 mL), CHCl₃:EtOAc (3:2) (250 mL), CHCl₃:EtOAc:MeOH (3:3:1) (525 mL), CHCl₃:EtOAC:MeOH (3:3:2) (900 mL), CHCL₃:MeOH (3:2) (750 mL), and MeOH (500 mL) successively. The anti-inflammatory bioactivity was found to be present in sub-fraction I.2-2, a fraction eluted with CHCl₃:EtOAC:MeOH (3:3:1). The I.2-2 sub-fraction was purified using a conditioned normal phase SPE (solid phase extraction) cartridge (Alltech; silica, 200 mg/4.0 mL, 50/P; particle size 50 μm and average pore size 60 å). Gradient elution consisting of CH₂Cl₂ (20 mL), CHCl₃ (20 mL), and MeOH (10 mL) was applied. The pure bioactive compound was eluted with CHCl₃ (20 mL).

Figure 1.  Acid–base extraction procedure adopted from Harborne (1998) to fractionate the methanol crude extract of the aerial parts of C. ainetensis.

Spectroscopic measurements

One-dimensional (¹H and ¹³C nuclear magnetic resonance (NMR) as well as DEPT (distortionless enhancement by polarization transfer) spectra and 2D NMR experiments were recorded using a Bruker 300 MHz spectrometer in deuterated chloroform (CDCl₃). Chemical shifts were reported in δ (ppm) values relative to tetramethylsilane (TMS). Gas chromatography (GC) analysis was performed using a Hewlett-Packard 6890 gas chromatograph equipped with HP-5 capillary column (30 m long, 250 μm i.d, and 0.25 μm film thickness). Helium was used as a carrier gas at a flow rate of 1 mL/min. The maximum temperature was 350°C, and the column was heated from 35°C to 290°C. The injector temperature was set at 300°C in a splitless mode. Results were recorded as percent of total peak areas. The mass spectrometer employed in the GC-mass spectroscopy (MS) analysis was a Hewlett-Packard 7972 series mass selective detector in the electron impact (EI) ionization mode (70 eV). Fourier transform infrared (FTIR) spectra were recorded on a Nicolet AVATAR 360 FTIR spectrometer. Ultraviolet-visible (UV-Vis) spectra were measured in methanol using a Jasco V-570 UV/VIS/NIR spectrophotometer.

SCp2 cell culture and treatment

SCp2 mouse mammary epithelial cells, sub-clones of CID-9 cells (Desprez et al., 1993), were seeded at 1 × 10⁶ cells/well (six-well tissue culture plates) in DMEM/F12 with 5% heat inactivated FBS, insulin (5 μg/mL), and gentamycin (50 μg/mL) in a humidified incubator (95% air, 5% CO₂) at 37°C. Twenty-four hours after plating, cells were washed three times with PBS and FBS-free DMEM/F12 medium, containing insulin (5 μg/mL), hydrocortisone (1 μg/mL), ovine prolactin (3 μg/mL), and gentamycin (50 μg/mL), and cells were incubated until confluent for 2–3 days. The medium was then supplemented with 1% FBS, and cells were treated with whole plant extracts or with I.1, I.2, I.3, I.4, and the resulting fractions from I.2, namely I.2-1–I.2-6, in the absence or presence of ET. Endotoxin, at 10 μg/mL, was added to the cells in culture 30 min post-treatment with plant extracts, while control cells were not treated with ET and/or plant extract as indicated. All treatments were performed in triplicate. Samples were collected 9 h post-ET treatment in the presence of “complete” protease inhibitors (final concentration of 40 μL of one tablet dissolved in 2 mL water per 1 mL of sample) and were stored at −80°C for further use.

Enzyme-linked immunosorbent assay (ELISA)

Interleukin-6 (IL-6) levels in the medium samples were measured using two-site (sandwich) ELISA CytoSets (BioSource International, Inc., USA) as described before (Safieh-Garabedian et al., 2004) and according to the manufacturer’s instructions. Enhanced binding 96-well plates (Cliniplate EB; ThermoLabsystems, Helsinki, Finland) were used in all ELISA assays. Samples were assayed in triplicate from duplicate experiments.

Mode-K cell culture and treatments

Murine intestinal epithelial cells, Mode-K cells, were maintained in DMEM (low glucose) containing 10% FBS, 1000 mg/L glucose, 10 mM sodium pyruvate, and non-essential amino acids. At 70−80% confluency, cells were detached by trypsinization and replated on tissue culture plastic flasks for maintenance (1:6 or 1:8 ratio of transfer) or were used for different experiments. For IL-1 treatment, cells were plated in growth media. Twenty-four hours later, the growth medium was replaced by an incomplete medium without FBS, and on the third day, cells were treated with 10 ng/mL IL-1 with or without C. ainetensis pure compound at different concentrations and for different time periods.

Trypan blue exclusion assay

To test for toxicity of the pure compound, Mode-K cells were grown in 12-well plates until 70–80% confluency. They were then treated with different concentrations of the C. ainetensis pure compound at different time points and different concentrations. At the time of harvesting, cells were washed and then trypsinized and added back to the wash suspension. Aliquots of 50 μL of the suspension were added to 50 μL of Trypan blue dye, and cells that did not take up the dye were counted as live cells and the percentage calculated as a percentage of the total number of cells present.

Preparation of homogenate and Western blotting

The effect of IL-1 and/or C. ainetensis pure compound treatment on the protein level of the enzyme cyclooxygenase was studied using Western blotting. Cells were washed twice in PBS, scraped in 2 × electrophoresis sample buffer (SB) containing Complete Protease Inhibitor Cocktail (40 μL/mL), 0.25 M Tris-HCl (pH 6.8), 4% w/v SDS, 20% w/w glycerol, and 0.1% bromophenol blue. Samples were boiled for 5 min and centrifuged, and the supernatant collected and stored at −80°C until further use. Total protein extracts were run on a 12% SDS-polyacrylamide gel at room temperature. After electrophoresis, proteins were transferred at 4°C overnight to PVDF membranes in a transfer buffer containing glycine, Tris-base, 10% SDS, and methanol (Tris-buffered saline (TBS) buffer). Following transfer, membranes were washed once with TTBS wash buffer (1 × TBS buffer containing 0.1% Tween 20) and then blocked for 1 h in 5% non-fat dry milk at room temperature. Antibodies were then added in TTBS and incubated for 2 h at room temperature. Unbound antibodies were washed three times with TTBS, after which horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G (IgG; at 1:5000 dilution) was added and membrane incubated for 60 min. Membranes were washed and incubated with luminal reagents at room temperature for 1 min and directly exposed to autoradiography.

Results and discussion

Structure determination of the isolated guaianolide

The bio-guided fractionation procedure illustrated in Figure 2i was followed in order to isolate and identify the guaianolide, salograviolide A (Figure 2ii). Spectroscopic measurements confirmed the reported data on salograviolide A, which was previously isolated and identified in C. salonitana Vis. (Daniewski et al., 1992) and C. nicolai Baldacci (Vajs et al., 1999). In addition, the position and stereochemistry of the hydroxyl groups were in agreement with the values of the coupling constants (J_8,9 = 8.06 Hz trans coupling; J_8,7 = 10.38 Hz; J_6,7 = 9.37 Hz) determined by Daniewski et al. (1992).

Figure 2.  (i) Diagram summarizing the fractionation of C. ainetensis extract. “A” indicates suppression of interleukin-6 (IL-6) production and “NA” indicates marginal or no suppression of IL-6 production by endotoxin (ET)-treated SCp2 cells. (ii) The suggested structure of the guaianolide, salograviolide A.

Anti-inflammatory properties of isolated guaianolide

ET-treated SCp2 cells, derived from mammary heterocellular CID-9, were used as a cell model of mastisis in vitro (Desprez et al., 1993; Schmidhauser et al., 1990). Interleukin-6 (IL-6), an inflammatory cytokine (Kariko et al., 2004) whose levels are increased significantly in ET-treated SCp2 in a concentration- and time-dependent manner, has been used previously in our laboratory as a marker for ET-induced inflammation both in vivo (Safieh-Garabedian et al., 2004) and in vitro (Safieh-Garabedian et al., 2004). Prior to methanol extract studies, crude water extracts from C. ainetensis were shown to downregulate, in vitro, ET-induced IL-6 and gelatinase levels in mammary CID-9 cells (parent strain of the SCp2 cells) and SCp2 cells, as well as suppress, in vivo, ET-induced paw edema and hyperalgesia (Talhouk et al., 2008). In this work, the ability of the methanol extract and subsequent fractions of C. ainetensis fractions to modulate IL-6 levels produced by ET-treated SCp2 cells was assessed by measuring the levels of IL-6 at 9 h following ET treatment in the presence of different concentrations of plant fractions (100 μg/mL for fractions I.1–I.4 and 16 μg/mL for fractions I.2-1–I.2-6). Fraction I.1, consisting mostly of waxes and fat residue, was not soluble in solvents compatible with cell culture media and was not assessed for its bioactivity. Out of fractions I.2, I.3, and I.4 that were dried and then reconstituted in either ethanol (EtOH) or diemthylsulfoxide (DMSO), only fraction I.2 suppressed ET-induced IL-6 levels (Figure 3A) at 100 μg/mL. This suppression was found to be concentration dependent with approximately 40% suppression at 50 μg/mL and no observed effect at 25 and 12.5 μg/mL (Figure 3B). Further fractionation of I.2 yielded subfractions I.2-1–I.2-6 which were also bioassayed for their anti-inflammatory activities. Out of these six sub-fractions, only I.2-2 showed anti-inflammatory effects (Figure 3C), an activity which was maintained when salograviolide A (isolate I.2-2) was purified and identified. This effect was concentration dependent and was observed at a concentration as low as 1 μg/mL (Figure 3D).

Figure 3.  Suppression of IL-6 production in ET-treated SCp2 cells by bioactive fractions of C. ainetensis: (A) Effect of I.2, I.3, and I.4 fractions (100 μg/mL) of C. ainetensis dissolved in ethanol (EtOH) or dimethylsulfoxide (DMSO) on IL-6 production (E, EtOH; D, DMSO). (B) Dose-dependent inhibition of IL-6 production by fraction I.2 at 100, 50, 25, and 12.5 μg/mL. (C) Effect of sub-fractions (I.2-2–I.2-6) at 16 μg/mL on IL-6 production. (D) Concentration-dependent effect of salograviolide A, purified fraction I.2-2, at 0.1, 0.5, 1, 2, 4, 6, and 8 μg/mL on IL-6 production. □ ▪ −ET, +ET, respectively. 100% Ctrl in (A) corresponds to 565 pg/mL of IL-6. Values in (A)–(D) were normalized to their own control values expressed as % control (Ctrl). Error bars show standard deviation of duplicate experiments.

Salograviolide A (isolate I.2-2) was tested on IL-1-induced cyclooxygenase-2 (COX-2) synthesis in intestinal epithelial cells. Interleukin-1 (IL-1), a proinflammatory cytokine, is produced by a variety of cell types in response to inflammatory stimulus or injury. In the intestine, increased IL-1 production has been reported in the mucosa of patients with inflammatory bowel disease (IBD) and in intestinal tissue of animal models of IBD, suggesting that IL-1 plays a role in the pathophysiology of IBD (Cominelli et al., 1989; Mahida et al., 1989). IL-1 is a major player in many inflammatory responses, including the induction of arachidonic acid mobilization and metabolism to prostaglandins (PGs) and other eicosanoids, a hallmark of inflammatory response. The cyclooxygenase (COX) enzyme converts arachidonic acid to prostaglandins and thromboxanes and exists in at least two isoforms, a constitutive isoform COX-1 and an inducible form COX-2. COX-1 represents most of the activity under basal conditions while COX-2 appears to be the predominant isoform present under stimulatory conditions, as in the case of inflammation. We have previously shown that IL-1 treatment of murine intestinal epithelial cells, Mode-K cells, induces COX-2 expression to maximal levels at 4–6 h following IL-1 treatment (Homaidan et al., 2001, 2002) without affecting COX-1 levels, suggesting the use of this cell system as a model of inflammation in vitro. Salograviolide A added 2 h prior to the addition of IL-1 to Mode-K cells caused a significant decrease in COX-2 levels in a concentration-dependent manner (Figure 4), while COX-1 levels, as expected, were not affected by such treatment (data not shown). All working concentrations of the plant extract and its sub-fractions did not affect cell viability.

Figure 4.  Modulation of cyclooxygenase-2 (COX-2) expression by salograviolide A. Western blot analysis demonstrates that 4 and 8 μg/mL of salograviolide A applied for 2 h to IL-1-treated Mode-K cells decreases COX-2 expression. β-Actin demonstrates equal loading (TF: salograviolide A).

In general, the structural requirement of α-methylidene-γ-butyrolactone to mediate its anti-inflammatory effects has been proposed to be the conjugated O=C—C=CH₂. This structural moiety acts as an alkylating agent by Michael-type addition with biological nucleophiles (Rüngeler et al., 1999) such as thiols or amines, which react by nucleophilic addition to the carbonyl group. As a result of this chemical reaction, the α-methylidene-γ-butyrolactone moiety tends to react non-specifically, especially at high concentrations. Furthermore, it has been shown that some sesquiterpene lactones have maintained activity after reducing their unsaturated γ-lactone moiety due to their selective non-covalent inhibition of proteins (Blanco et al., 2001), while others have lost their activity upon reduction (Kwok et al., 2001) and some have increased their activity by adding more than one Michael acceptor site (Lyss et al., 1998). In reference to salograviolide A, the presence of three exocyclic methylene groups, in addition to the unsaturated γ-lactone, could increase its reactivity and selectivity toward nucleophiles in proteins. Hence, derivatives to reduce the exomethylene functional groups situated at C₄, C₁₀, and C₁₁ and their biological effects are under study.

Conclusion

Secondary metabolites of the genus Centaurea have been reported to have a multitude of biological activities including antioxidant, antifungal, antimicrobial, hypoglycemic, and anti-tumor activities. After following an anti-inflammatory guided fractionation method of extraction and isolation, salograviolide A from C. ainetensis was identified. This is the first report that describes the anti-inflammatory activity of salograviolide A. The anti-inflammatory activity noted in ET-treated SCp2 cells was concentration dependent, and suppression of IL-6 production was observed at concentrations of 1 μg/mL. In IL-1-treated Mode-K cells, salograviolide A (at 2 μg/mL) downregulated IL-1-induced COX-2 levels. The results of this study serve to define the composition of the bioactive extracts from C. ainetensis, and establish the preliminary steps in order to investigate the mechanism of the anti-inflammatory actions of salograviolide A in cell and animal models of inflammation.

Declaration of interest: This work was supported by grants from HITECH, FZE, Dubai, UAE, and the research was conducted at the IBSAR Center.