Abstract
The aim of this study was to investigate the effects of cucurbitacin R on an experimental model of adjuvant-induced arthritis in rats. The treatment of arthritic rats with cucurbitacin R (1 mg/kg p.o. daily) modified the evolution of the clinical symptoms, whereas the histopathology of paws demonstrated a reduction in the signs of arthritis. Compared with the control group, radiography of the tibiotarsal joints of cucurbitacin R-treated rats showed a decrease in joint damage and soft tissue swelling of the footpad. The in vivo study of the expression of proinflammatory enzymes (nitric-oxide synthase-2 and cyclooxygenase-2) with the aid of the Western blot technique, and that of tumor necrosis factor-α (TNF-α) and prostaglandin E2 by means of enzyme-linked immunosorbent assays demonstrated a clear decrease in both the enzymes and the mediators in paw homogenates. The analysis for prostaglandin E2, nitric oxide, and TNF-α production in RAW 264.7 macrophages, as well as that for TNF-α in human lymphocytes, indicated a reduction of all mediators. The expression of cyclooxygenase-2 was not modified in RAW 264.7 macrophages, whereas the expression of nitric-oxide synthase-2 was clearly diminished. Moreover, cucurbitacin R was found to inhibit signal transducer and activator of transcription 3 activation in the lymphocytes of both healthy and arthritic men. These experimental data on the chronic model, together with previously reported activity on acute and subchronic experimental models, justify the anti-inflammatory activity of cucurbitacin R and provide further evidence for the therapeutic potential of a group of natural products as anti-inflammatory agents.
Arthritis is a chronic disease that affects several parts of the joints such as the cartilage, synovium, tendons, and muscles. Rheumatoid arthritis is a specific type of this disease characterized by chronic inflammation of the joints, with severe cartilage and bone damage. The precise causes are unknown, but genetics, infectious agents, environment, and hormonal effects have all been implicated. Treatments are focused on the reduction of pain, inflammation, and joint damage. The principal pharmacological agents are nonsteroidal anti-inflammatory drugs, disease-modifying antirheumatic drugs, glucocorticoids, and specific inhibitors of the mediator response such as the tumor necrosis factor-α (TNF-α) antibody infliximab. Adjuvant arthritis is an experimental model unique to rats that is widely used for studying the physiology, biochemistry, and pharmacology of inflammation and also as a model of cell-mediated autoimmnune disease, human arthritis, and chronic pain. Different parameters can be evaluated, such as weight evolution, paw swelling, histology, immunohistochemistry, and radiographic evaluation (Taurog et al., 1988). The origin and development are complex, with the implication of different inflammatory cells and cytokines. Thus, when the inflammatory cells migrate into the joint space between bones, there is inflammation of the membranes and damage to the bone and cartilage, with cytokines being released to transmit the signal. Neutrophils, macrophages, and lymphocytes all participate in the development and pathogenesis of arthritis (Solomon et al., 2005). Neutrophils generate reactive oxygen species, releasing hydrolytic enzymes, such as elastase (Bombini et al., 2004). Activated T lymphocytes increase the levels of TNF-α (Skapenko et al., 2005), and activated macrophages generate oxygen and nitrogen reactive species, eicosanoids through cyclooxygenase-2, nitric oxide through the inducible form of nitric-oxide synthase, and cytokines such as TNF-α (Park et al., 2004a,b). All these enzymes and mediators are involved in the development of rheumatoid arthritis, stimulating the leukocyte migration, inflammation in the joint, and tissue damage.
Superoxide anion, which is formed by inflammatory cells in chronic inflammation, is generally metabolized by the endogenous superoxide dismutase. However, when production of the former is extremely high, it overwhelms the action of the latter. In such situations, superoxide anion can mediate cell damage, increase the formation of chemotactic factors such as leukotriene B4, and stimulate recruitment of neutrophils at sites of inflammation that could be aggravated (Salvemini et al., 2003). Because neutrophil elastase is involved in the destruction and inflammation of tissue in different diseases, including rheumatoid arthritis, the development of new inhibitors of this protease is an important research objective (Tremblay et al., 2003).
Nociception involves a complex interaction of peripheral and central nervous system structures extending from the skin, the viscera, and the musculoskeletal tissues to the cerebral cortex. Various cells, mediators, and receptors are implicated in this process; therefore, the anti-inflammatory drugs currently used for treatment act by reducing the production and release of these mediators and by modifying the access to their targets. Ribeiro et al. (2000) demonstrated that the nociceptive response induced by acetic acid depends on peritoneal resident macrophages and mast cells through the release of TNF-α, interleukin-1β, and interleukin-8.
Tayuya is a medicinal plant used in the folk medicines of Brazil, Peru, and Colombia as an analgesic, anti-inflammatory, and anti-rheumatic agent (Ríos et al., 1990). Cucurbitacin R, isolated from tayuya roots, has shown anti-inflammatory activity in different experimental models of acute mouse ear edema, acute mouse paw edema, and mouse ear inflammation (Recio et al., 2004). Moreover, Park et al. (2001, 2004a) demonstrated that cucurbitacin R (called dihydrocucurbitacin D in these articles) inhibited nitric oxide generation from murine macrophages by blocking the activation of nuclear factor-κB, which is essential for the transcriptional activation of nitric-oxide synthase induction. In 1999, Panossian et al. reported that cucurbitacin R diglucoside moderately stimulated the adrenal cortex to produce a slight increase in corticoid secretion, thereby justifying its pharmacological properties. However, our previous results demonstrated that cucurbitacin R did not modify its anti-inflammatory properties when administered together with mifepristone, which implied that the glucocorticoid receptor is not involved in the mechanism of action of cucurbitacin R. However, when the compound was administered together with cycloheximide, the anti-inflammatory activity was clearly reduced (Recio et al., 2004). Recently, Escandell et al. (2006) demonstrated the anti-arthritic effect of a closely related compound, dihydrocucurbitacin B, which reduced the edema and the tissue and bone damage. In the present study, we report on the effects of cucurbitacin R on a chronic model of inflammation (adjuvant arthritis) in rats, the effects on cartilage and bone damage, and the effect on the principal mediators implicated in this experimental process.
Materials and Methods
Reagents and Chemicals
Cucurbitacin R (Fig. 1) was previously isolated from Cayaponia tayuya roots (Recio et al., 2004). Chemicals were obtained from Panreac (Barcelona, Spain) and Merck Biosciences (Darmstadt, Germany). Biochemical reagents were from Sigma-Aldrich (St. Louis, MO). Antibodies were purchased from Cayman Chemical (Ann Arbor, MI) and Santa Cruz Biotechnology (Santa Cruz, CA). Biotin-conjugated goat anti-rabbit immunoglobulin G and avidin-biotin peroxidase complex (LSAB 2 System-HRP) were from Dako (Glostrup, Denmark). The kits for enzyme immunoassay of cytokines were from eBioscience (San Diego, CA). Culture material and reagents were from Invitrogen (Langley, OK), Sarstedt (Nümbrecht, Germany), and NUNC A/S (Roskilde, Denmark). Complete mini EDTA-free protease inhibitor cocktail was purchased from Roche Diagnostics (Mannheim, Germany).
Animals
Groups of female Lewis rats weighing 200 to 240 g (7 weeks old) and groups of Swiss female mice weighing 25 to 30 g (8 weeks old) from Harlan Interfauna Iberica (Barcelona, Spain) were used for arthritis and analgesia, respectively. All animals were fed a standard diet ad libitum. Housing conditions and all in vivo experiments were approved by the institutional ethical committee of the Faculty of Pharmacy according to the guidelines established by the European Union on Animal Care (CEE Council 86/609).
Adjuvant Arthritis
Adjuvant arthritis was elicited in female Lewis rats by injecting Mycobacterium butyricum (0.1 ml, 10 mg/ml) from Difco (Detroit, MI) in mineral oil into the base of the tail. Paw volumes were measured at the beginning of the experiment with a plethysmometer (Ugo Basile, Comerio, Italy). The initial edema was evaluated by measuring the volume of both paws at day 16. Animals with paw volumes 1.1 ml larger than normal paws were then randomized into treatment groups. Cucurbitacin R (1 and 0.25 mg/kg), the reference drug ibuprofen (10 mg/kg; Sigma-Aldrich) and vehicle as control (olive oil; Carbonell, Córdoba, Spain) were administered p.o. once daily on days 17 to 23, and the paw edemas were measured on the same days. Edema was expressed as the increase in paw volume due to arthritis, and the percentage of inhibition was expressed as the reduction in volume with respect to the control group. The rats were also weighed each day and weight variations were compared with those of the control group.
Clinical Assessment and Radiographic Studies of Arthritis
The rats were previously anesthetized with halothane (Sigma-Aldrich) and placed on a radiographic box at a distance of 90 cm from the X-ray source. Radiographic analysis of normal and arthritic hind paws was performed with an X-ray machine (Phillips X12, Munich, Germany) with a 40 kW exposition for 0.01 s. The following radiograph parameters were considered: inflammation, osteolysis, articular cartilage affectation, and osteophyte formation, with the following criteria and scores: 0, no damage; 1, mild; 2, moderate; and 3, severe.
Histological Evaluation and Immunohistochemical Localization of Cyclooxygenase-2
Animals were previously anesthetized and then sacrificed by decapitation, and arthritic paws were amputated above the ankle and were placed in 4% formalin (Panreac). The paws were then trimmed, placed in decalcifying solution for 24 h, embedded in paraffin (Panreac), sectioned at 4 μm, stained with trichromic Masson (prepared with products from Sigma-Aldrich), and studied using light microscopy (Dialux 22; Leitz, Wetzlar, Germany).
Cyclooxygenase-2 was determined by immunohistochemical analysis. At day 24, the organs of the joint were trimmed and placed in decalcifying solution for 24 h, and 8-μm sections were prepared from paraffin-embedded tissues. After deparaffinization, endogenous peroxidase was quenched with 0.3% H2O2 (Panreac) in 60% methanol (Panreac) for 30 min. The sections were permeabilized with 0.1% Triton X-100 (Sigma-Aldrich) in phosphate-buffered saline (PBS) (Sigma-Aldrich) for 20 min. Nonspecific adsorption was minimized by incubating the section in 2% normal goat serum (Sigma-Aldrich) in PBS for 20 min. Endogenous biotin- or avidin-binding sites were blocked by sequential incubation for 15 min with avidin and biotin (LSAB2). The sections were then incubated overnight with primary anti-cyclooxygenase-2 (SC-1747; Santa Cruz Biotechnology) (1:500) with control solutions. Control included buffer alone or nonspecific purified rabbit immunoglobulin G (Sigma-Aldrich). Specific labeling was detected with a biotin-conjugated goat anti-rabbit immunoglobulin G (LSAB2) and avidin-biotin peroxidase complex (LSAB2).
Prostaglandin E2, TNF-α, and Western Blot Analysis in Paw and Stomach Homogenates
For the measurement of prostaglandin E2 in paw and stomach, the organs were homogenized with a Polytron (Kinematica AG, Lucerne, Switzerland) in 0.1 M phosphate buffer, pH 7.4, containing 1 mM EDTA (Sigma-Aldrich), and 10 μM indomethacin (Sigma-Aldrich). Prostaglandin E2 was isolated using a C18 SPE cartridge (LiChrolut; Merck Biosciences) and was determined by a specific enzyme immunoassay kit from Cayman Chemical, used according to the manufacturer's instructions.
TNF-α enzyme-linked immunosorbent assay and Western blot analysis of paws were performed with samples from control, ibuprofen and cucurbitacin R groups. Skin was removed from paws; they were then homogenized with a Polytron in 2.5 ml of 10 mM HEPES, 2 mM phenylmethylsulfonyl fluoride, 100 mM EDTA, 0.32 M sucrose, 1 mM dithiothreitol, 2 mg/ml aprotinin (all from Sigma Chemical), and complete mini EDTA-free protease inhibitor cocktail. The mixture was sonicated (three cycles in 10 s) in a Branson sonifier 150 (Branson Ultrasonics Corporation, Danbury, CT) and centrifuged (Eppendorf centrifuge 5810R; Eppendorf, Hamburg, Germany) at 10,000g for 20 min at 4°C. TNF-α was measured by enzyme immunoassay (eBioscience), used according to the manufacturer's instructions.
Protein in supernatant was measured by the Bradford method using bovine serum albumin (Sigma-Aldrich) as standard, and 30 μg of protein were loaded on 10% SDS-PAGE (products from Sigma Chemical) and transferred onto polyvinylidene difluoride membranes (GE Healthcare, Little Chalfont, Buckinghamshire, UK) for 90 min at 125 mA. Membranes were blocked in PBS-Tween 20 (Panreac) containing 5% w/v defatted milk (Santa Cruz Biotechnology). Membranes were incubated with anti-cyclooxygenase-2 polyclonal antiserum (1:1000; Cayman Chemical), anti-nitric-oxide synthase-2 antibody (1:1000; Cayman Chemical), and β-actin (1:10000; Sigma-Aldrich). Blots were washed and incubated with peroxidase-conjugated goat anti-rabbit immunoglobulin G (1:20,000 dilution; Sigma-Aldrich). The immunoreactive bands were visualized using an enhanced chemiluminescence system (GE Healthcare).
Analgesia
Cucurbitacin R (10 mg/kg) and ibuprofen (30 mg/kg) were administered p.o. 1 h, before the i.p. administration of 1% acetic acid (Panreac) solution (0.1 ml). Control animals received vehicle in the same experimental conditions. Immediately after acetic acid injection, each animal was isolated in an individual cage to be observed during 20 min. The numbers of instances of writhing and stretching were recorded. Analgesic activity was expressed as the percentage of writhing reduction referred to the control group.
Determination of Cell Viability
The cytotoxicity of compounds on cells was performed by the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenil-tetrazolium bromide (MTT) assay. Human neutrophils, human lymphocytes, and murine RAW 264.7 macrophages were exposed to cucurbitacin R at a concentration of 100 μM in a microplate at the assayed time in the different experiments, and then 100 μl/well of a 0.5 mg/ml solution of MTT (Sigma-Aldrich) were added and the solution was incubated at 37°C until blue deposits were visible. The colored metabolite was dissolved in dimethyl sulfoxide. Absorbance was measured at 490 nm using a Labsystems Multiskan EX plate reader (Helsinki, Finland). Results were expressed in absolute absorbance readings; a decrease indicated a reduction in cell viability.
Preparation of Human Neutrophils for Elastase Release/Activity and Radical Superoxide Assays
Neutrophils were obtained from human blood buffy coats. The leukocytes present in the residual blood were isolated by sedimentation with 2% (w/v) dextran (Sigma-Aldrich) in 0.9% NaCl (Panreac) at room temperature. The supernatant was centrifuged at 300g for 10 min at 4°C. The rest of the erythrocytes were lysed by hypotonic treatment. The pellet was resuspended in PBS, and cells were purified by the Ficoll-Paque gradient density method (GE Healthcare). Neutrophils (95% purity) were resuspended in PBS containing 1.26 mM Ca2+ and 0.9 mM Mg2+ (Sigma-Aldrich).
Elastase Release Assay. Human neutrophils (2.5 × 106) were preincubated for 5 min with the test compound (100 μM) or vehicle. Then they were stimulated by the addition of 10 μM cytochalasin B (Sigma-Aldrich) and 10 nM N-formyl-l-methionyl-l-leucyl-l-phenylalanine (Sigma-Aldrich) for 10 min at 37°C; 15 min later the samples are placed on ice and centrifuged (0°C for 10 min at 3220g). Supernatant aliquots (200 μl) were mixed with 5 μl of 0.3% N-tert-butoxycarbonyl-l-alanine p-nitrophenyl ester (Sigma-Aldrich) in the 96-well microtiter plate and then were incubated at 37°C for 30 min. Elastase activity was detected using N-tert-butoxy-carbonyl-l-alanine p-nitrophenyl ester as substrate, and p-nitrophenol release was measured at 414 nm. α1-Antitrypsin (Sigma-Aldrich) was used as reference (90 μg/ml).
Elastase Activity Assay. Human neutrophils (20 × 106) were suspended in 8 ml of Ca2+/Mg2+ Hanks' balanced salt solution (Sigma-Aldrich) in a Falcon tube. Then 80 μl of 12-O-tetradecanoylphorbol 13-acetate (65 μg/ml; Sigma-Aldrich) were added, and the cells were incubated at 37°C for 30 min. The samples were centrifuged (0°C for 10 min at 3220g). Supernatant aliquots corresponding to 1.25 × 106 neutrophils were incubated for 15 min with 10 μl of the test compound to a final concentration of 100 μM before proceeding as in the release assay. The rest of the protocol was as described above.
Radical Superoxide Assay. Neutrophils (2.5 × 106), obtained as described above, were suspended in 500 μl of Hanks' balanced salt solution containing Ca2+ and Mg2+ and test compounds at different concentrations, and the mixture was incubated for 5 min at 37°C. Superoxide release was induced by addition of 5 μl of 12-O-tetradecanoylphorbol 13-acetate (final concentration 1 μM) and after a 10-min incubation at 37°C, the mixture was centrifuged at 3220g. The reaction was detected by nitro blue tetrazolium (100 μM) reduction. The precipitate was dissolved in 500 μl of dimethyl sulfoxide-HCl (95:5) in an ultrasonic bath (Branson sonifier 150) and measured using a Labsystem Multiskan EX plate reader at 560 nm.
Determination of TNF-α, Nitric Oxide and Prostaglandin E2 Production, and Nitric-Oxide Synthase-2 and Cyclooxygenase-2 Activity, in RAW 264.7 Macrophages
RAW 264.7 macrophages were cultured in Dulbecco's modified Eagle's medium containing 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, and 10% fetal bovine serum (all from Invitrogen). Cells were removed from the tissue culture flask using a cell scraper and resuspended to a final relation of 1 × 106 cells/ml.
RAW 264.7 macrophages (1 × 106 cells/ml) were coincubated in 96-well culture plates (200 μl) with 1 μg/ml lipopolysaccharide (Sigma-Aldrich) at 37°C for 24 h in the presence of test compound (100–25 μM) or vehicle. Nitrites were determined in culture supernatant by using Griess reagent (Sigma-Aldrich). Prostaglandin E2 production was determined in culture supernatant by a specific enzyme immunoassay kit from Cayman Chemical, used according to the manufacturer's instructions, and TNF-α was determined by a specific enzyme immunoassay kit.
RAW 264.7 macrophages (1 × 106 cells/ml) were stimulated with 1 μg/ml lipopolysaccharide in the absence of test compound. After 24 h, cells were washed, and fresh medium was supplemented with 0.5 mM l-arginine (Sigma-Aldrich) and 5 μM arachidonic acid (Sigma-Aldrich) and incubated for 2 h in presence of test compounds (100–25 μM). Supernatants were collected for the measurement of nitrite and prostaglandin E2 accumulation for the last 2 h. In this experiment, the metabolites were assessed as an index of nitric-oxide synthase-2 and cyclooxygenase-2 activity.
Determination of TNF-α Production in T Lymphocytes
Peripheral lymphocytes were obtained from the blood of both healthy and arthritic men; rheumatoid arthritis in the latter had been clinically diagnosed according to the criteria of the American Rheumatism Association (Arnett et al., 1988). Cells were isolated by the Ficoll-Paque gradient density method (GE Healthcare). T lymphocytes were isolated by depletion of adherent cells on plastic dishes (95% purity). Under sterile conditions, cells were resuspended to a concentration of 1 × 106 cells/ml in RPMI 1640 medium (Invitrogen), supplemented with 10% heat-inactivated fetal calf serum, 100 U/ml penicillin, and 100 μg/ml streptomycin. T lymphocytes were cultured with 25 μg/ml phytohemagglutinin (Sigma-Aldrich) alone or with cucurbitacin R at different concentrations (40–3 μM) for 24 h. The cell supernatants were collected and assayed for TNF-α by enzyme immunoassay.
Western Blot Analysis
Cellular lysates from RAW 264.7 macrophages (106 cells/well) incubated for 18 h with 1 μg/ml lipopolysaccharide and 10 μM dexamethasone (Sigma-Aldrich) or test compound at different concentrations (100–25 μM) were obtained with lysis buffer [1% Triton X-100, 1% deoxycholic acid, 20 mM NaCl, and 25 mM Tris, pH 7.4 (all from Sigma-Aldrich), and a complete mini EDTA-free protease inhibitor cocktail]. After centrifugation (10,000g, Eppendorf centrifuge 5810R), proteins were determined in the supernatants by means of the Bradford method. Fifty micrograms of protein were developed as described earlier under Prostaglandin E2, TNF-α, and Western Blot Analysis in Paw and Stomach Homogenates.
For the STAT3 assay, human lymphocytes from healthy and arthritic patients were treated for 4 h with vehicle or 50 μM cucurbitacin R and then stimulated with either 20 ng/ml interleukin-6 (Prepotech, London, UK) for 30 min or 25 μg/ml phytohemagglutinin for 2 h. The cells were then lysed with RIPA buffer (1% Triton X-100, 1% deoxycholic acid, 150 mM NaCl, 50 mM Tris, pH 7.4, and 0.1% sodium dodecyl sulfate), 2 mM sodium orthovanadate, and 25 mM NaF (all from Sigma-Aldrich), and a complete mini EDTA-free protease inhibitor cocktail. After centrifugation (10,000g), proteins were determined in the supernatants by means of the Bradford method. One hundred micrograms of protein were run on 7.5% SDS-PAGE gel (Santa Cruz Biotechnology). The protein was transferred to nitrocellulose and then blotted as described above for phospho-specific STAT3 (Santa Cruz Biotechnology).
Statistics
Data are expressed as means ± S.E.M. Statistical analysis was performed by one-way analysis of variance (ANOVA) followed by Dunnett's t test for multiple comparisons. For comparisons with the control group, values of P < 0.05 were considered significant. Inhibition percentages (%I) were calculated from the differences between drug-treated groups and control animals treated only with the inflammatory agent. IC50 and ED50 were calculated from the dose-response linear regression plots. Western blot quantification was carried out with Scion Image software (version 1.0.0.1; Scion Corporation, Frederick, MA).
Results
Evolution of Induced Arthritis in Rats. Arthritis was clearly developed in the immunized rats 2 weeks after challenge. Periarticular erythema, edema, and a reduction of paw function were observed as clear signs of the evolution of the inflammation, and a 100% incidence of adjuvant arthritis in the rats was observed by day 16. All rats experienced a dramatic increase in paw volume from 1.1 to 2.5 ml during the 17 days after challenge, after which time the paw volume of the control group stabilized until the end of the experiment. Cucurbitacin R (p.o. at 1 mg/kg for 7 days) significantly reduced hind paw swelling by 48%, whereas at 0.25 mg/kg a reduction of 10% was observed (not significant, data not shown). The reference drug ibuprofen (10 mg/kg) reduced the swelling by 58% under the same experimental conditions (Fig. 2A). The blank group showed no increase in hind paw volume during the experiment. There was a decrease in weight in all arthritic animals, but although the weight in all groups stabilized during the experiment, the weight loss was lower in the group treated with cucurbitacin R (Fig. 2B). Still, as mentioned above, cucurbitacin R did not significantly modify the edema at 0.25 mg/kg with respect to the control group.
Radiography of the tibiotarsal joint of rats demonstrated severe inflammation in the control group with severe periarticular inflammation, osteolysis of the tarsal bone, calcification of the Achilles tendon, effects in the distal tibia and severe alteration of proximal metatarsal bones, bone resorption, and joint erosion (Fig. 3A), whereas the radiography of the tibiotarsal joint of the cucurbitacin R-treated rats indicated a decrease in joint damage as well as in soft tissue swelling of the footpad and only light osteolysis of tarsalmetatarsal zone with light calcification (Fig. 3C). Ibuprofen-treated rats show radiographic characteristics similar to those of the cucurbitacin R-treated group (Fig. 3B).
Histopathological Study Production of Mediators in Vivo. The histopathological study of paws obtained from the control group showed characteristic signs of severe skin inflammation (Fig. 4A) and arthritis, with massive mixed infiltration (neutrophils, macrophages, and lymphocytes), lesions with marked edema and severe abscesses, severe erosion of bone and cartilage (Fig. 4B), and bone destruction (Fig. 4C), articular inflammation, pannus, and large cysts with abundant granulation tissue. More than 50% of the bone surfaces were affected; bone regeneration by osteoid was >75% (Fig. 4C), and the thickness of periosteal regeneration was >500 μm. The histological analysis of joint sections obtained from cucurbitacin R-treated rats showed that the degree of arthritis was clearly attenuated (Fig. 5C) compared with that in the control group (Fig. 5B). In animals treated with ibuprofen, the histological signs of arthritis decreased (Fig. 5D), whereas the blank group showed no characteristic arthritic lesions (Fig. 5A).
Immunohistochemical analysis of joint sections obtained from adjuvant arthritic rats revealed positive staining for cyclooxygenase-2 (Fig. 6A). In contrast, this staining was reduced in the joints of cucurbitacin R (Fig. 6B)- and ibuprofen-treated rats (Fig. 6C). Moreover, cucurbitacin R inhibited the in vivo production of prostaglandin E2 by 94% and reduced both the nitric-oxide synthase-2 and cyclooxygenase-2 expression in paw homogenates as demonstrated with Western blot analysis (Fig. 7A). Prostaglandin E2 production in the stomach homogenates was also reduced (Fig. 7B), although there was no visible damage in the tissue (data not shown). Cucurbitacin R inhibited TNF-α production in arthritic paws by 76%, and ibuprofen reduced it by 68% (Fig. 7C).
Analgesia, Elastase Release and Elastase Activity, and Effect on Superoxide Radical. Cucurbitacin R administered orally at 10 mg/kg showed significant analgesic activity, with a 47% reduction in writhing compared with the control group (p < 0.05). The group treated with ibuprofen (30 mg/kg) showed a 50% inhibition (p < 0.05). In addition, cucurbitacin R had no effect on superoxide radical (data not shown). Cucurbitacin R at 100 μM inhibited elastase release by 50% (p < 0.05) but had no effect at lower concentrations, nor did it modify the enzyme activity.
Cell Viability. As assessed by the mitochondrial reduction of MTT after 24 h, cucurbitacin R did not affect the cellular viability of human neutrophils, human lymphocytes, or murine RAW 264.7 cells. This finding indicates that the compound is not toxic at doses of ≤100 μM.
Nitrite, Prostaglandin E2 and TNF-α Production in RAW 264.7 Macrophages and Lymphocytes. During the induction phase, cucurbitacin R inhibited nitrite production by 42% at a final concentration of 100 μM (24 h), by 22% at 30 μM, and by 14% at 10 μM. The positive control, l-NAME (Sigma-Aldrich), a specific inhibitor of nitric-oxide synthase-2, inhibited nitrite production by 98% at 1 mM, whereas dexamethasone inhibited this production by 37% at 10 μM (Fig. 8A). In addition, cucurbitacin R inhibited prostaglandin E2 production by 53%, whereas dexamethasone at 10 μM reduced the prostaglandin E2 production by only 43% (Fig. 8B).
When the enzymes were induced before to the treatment (postinduction phase assay), cucurbitacin R and dexamethasone had no effect on nitrite production, which implies that neither of them has a direct effect on the enzyme. However, l-NAME inhibited nitrite production by 42% because of direct inhibition of the enzyme (Fig. 8A). In the cyclooxygenase-2 postinduction phase assay, cucurbitacin R at 100 μM inhibited the enzyme activity by 60%, whereas indomethacin at 10 μM reduced it by 81% (Fig. 8B). In this case, both compounds directly modify the activity of the enzyme cyclooxygenase-2.
These findings were corroborated after an assay involving the coincubation of cucurbitacin R at different concentrations on lipopolysaccharide-stimulated RAW 264.7 macrophages to assess the possible effects of cucurbitacin R on the expression of nitric-oxide synthase-2 or cyclooxygenase-2 (Fig. 8C). The results clearly indicate that in the presence of cucurbitacin R at 100 and 50 μM, nitric-oxide synthase-2 expression was inhibited by 57 and 40%, respectively. Cyclooxygenase-2 induction, however, was not affected. As was expected, the reference compound dexamethasone inhibited the expression of both enzymes.
Cucurbitacin R inhibited TNF-α production in a concentration-dependent manner with an IC50 value of 65 μM (52–78, r2 = 0.9866) in RAW 264.7 macrophages (Fig. 9A), 29 μM (26.0–36.5, r2 = 0.9705) in healthy human lymphocytes (Fig. 9B), and 7 μM (4.0–8.8, r2 = 0.9612) in lymphocytes from arthritic patients (Fig. 9C). The differences in potency between human and mouse cells may be attributable to the facts that the tests were performed on different cells from different species and also that the stimuli used were not the same. In human lymphocytes, they are due to the difference in TNF-α production and the sensibility of cells from arthritic and healthy patients.
STAT3 Activation in Human Lymphocytes from Healthy Men and Arthritic Patients. Finally, cucurbitacin R inhibited the activation of STAT3 by interleukin-6. It should be noted that there were clear differences between human lymphocytes obtained from healthy men and those from arthritic patients. For example, whereas in the latter group the STAT3 was activated, this was not the case in the lymphocytes from healthy donors. In both instances, however, cucurbitacin R was found to inhibit the activation of STAT3 induced by interleukin-6 (Fig. 10).
Discussion
TNF-α and interleukin-1β play a relevant role in the pathogenesis of adjuvant arthritis, but other mediators such as interleukin-6, interleukin-15, interleukin-18, and leukotriene B4 are also implicated in the process, with some of them playing an important role in neutrophil recruitment during immune inflammation. TNF-α, which is involved in inflammation, differentiation, and proliferation of T and B cells, and bone resorption, is the primary agent in the inflammatory process (Rioja et al., 2004), whereas interleukin-1β is responsible for the destruction of cartilage and bone (Cuzzocrea et al., 2000). Blocking TNF-α suppresses the inflammation and ameliorates cartilage destruction (Cuzzocrea et al., 2000), as this cytokine not only plays a relevant role in leukocyte recruitment to the articulations (Issecutz et al., 1994) but also regulates nitric-oxide synthase-2 and cyclooxygenase-2 expression in the synovial tissue and cartilage of arthritic rats (Amin et al., 1999). This is significant as the metabolites of these enzymes, nitric oxide and prostaglandin E2, have an essential function in the development of the inflammatory process (Fahmi, 2004).
Besides TNF-α, interleukin-6 also plays an important role in arthritis. In fact, a recent series of in vivo studies using models of experimental arthritis have established the role of interleukin-6 in joint destruction, leukocyte recruitment, apoptosis, and T cell activation (Scheller et al., 2006). For its part, interleukin-15 is implicated in the recruitment and activation of T lymphocytes and neutrophils; its neutralization has been found to produce an amelioration of inflammation and articular destruction in murine collagen-induced arthritis (Baslund et al., 2005). Interleukin-18 has the ability to induce the production of tumor necrosis factor-α and interleukin-1β in mononuclear cells and to initiate a cytokine cascade with concomitant expression of proinflammatory markers such as chemokines, nitric oxide, adhesion molecules, and matrix metalloproteinase-9 (Muhl and Pfeilschifter, 2004). The inflammatory synovial fluid thus predominantly contains neutrophils, whereas, in addition to cytokines, leukotrienes are among the inflammatory mediators expressed in the inflamed joint (Chen et al., 2006). In patients with rheumatoid arthritis, elevated levels of leukotriene B4 correlate with disease severity, whereas leukocytes in the synovial fluid highly express BLT1 (Kim et al., 2006). By taking all this into consideration and in light of our results, the reduction in inflammation as well as that in cartilage destruction brought about by cucurbitacin R is probably due to decreases in TNF-α production, the induction of both nitric-oxide synthase-2 and cyclooxygenase-2 observed in the damaged tissue, and a reduction in cell infiltration but not to a decrease in leukotriene B4 production, as was demonstrated in a previous article (Recio et al., 2004).
Cucurbitacins are usually reported as cytotoxic compounds, but there are clear differences between the toxicity and activity of cucurbitacins, depending on the pattern of substitution. Only a few studies on the in vivo toxicity of cucurbitacins have been described (Ríos et al., 2005). One conclusion was that their range of toxicity was between 2 and 12.5 mg/kg. However, in previous tests involving the in vivo toxicity of a mixture of cucurbitacins, including cucurbitacin R, we obtained an LD50 of 375 mg/kg p.o. and 67 mg/kg i.p. (Ríos et al., 1990). Musza et al. (1994) demonstrated that the presence of an acetyl group at C-25 of the chain clearly increases the toxicity, whereas Oh et al. (2002) showed that the presence of a double bond at C-23 also produces an increase in the cytotoxicity of these compounds. The fact that these structural features are implicated in the pharmacological activity was further bolstered by work carried out by Jayaprakasam et al. (2003), who demonstrated the selective inhibition of cucurbitacins with the acetyl group at C-25 against cyclooxygenase-2 versus cyclooxygenase-1 while showing that the same pattern of inhibition did not hold in the case of deacetyl derivatives such as cucurbitacin R. However, in our work this cucurbitacin (without an acetyl group) inhibited the cyclooxygenase-2 induction in paws and the prostaglandin E2 production in both paws and stomach, which implies an inhibition of the activity of both cyclooxygenase-1 and cyclooxygenase-2. The results obtained for ibuprofen-treated rats were similar in stomach, but activity in paws was lower at higher doses, which could imply a similar anti-inflammatory effect with lower gastric side effect at the same dose. In addition, in tests on the effects on isolated cells, cucurbitacin R at 100 μM reduced the prostaglandin E2 production when administered before and after the induction of the enzyme, a fact that, together with the Western blot analysis in RAW 264.7 cells of cyclooxygenase-2, demonstrates the inhibition of the activity of the enzyme.
Our aim in studying the effects of cucurbitacin R on a model of experimental analgesia was thus to see the in vivo response to an algogenic stimulus. Whereas previous experiments using doses of 5 mg/kg cucurbitacin R produced no effect on the writhing induced by acetic acid in mice, in the present investigation we obtained a significant reduction of writhing when a higher dose of 10 mg/kg was used. This result corroborates findings previously reported by Peters et al. (1997), who tested a mixture of cucurbitacins against the acetic acid-induced algesic response in mice. These same authors (Peters et al., 2003) demonstrated that inhibition of nitric oxide formation and cyclooxygenase activity is implicated in the analgesic activity of the cucurbitacins studied. Moreover, they demonstrated the peripheral effects of the treatment because cucurbitacins abolished the abdominal contortions in the writhing test but had no effect in the hot plate and rotarod test. Ribeiro et al. (2000) demonstrated that the nociceptive response induced by acetic acid depends on the peritoneal resident macrophages and mast cells through the release of TNF-α, interleukin-1β, and interleukin-8. As for TNF-α, it has been found to play a pivotal role in the genesis of the nociceptive writhing response in mice. This effect should thus be diminished if TNF-α production is inhibited, a theory that may explain the in vivo effect of cucurbitacin R on the writhing test in mice.
The role of nitric oxide in the pathogenic damage of induced arthritis is subject to some controversy. It has been shown, for example, that the induction of nitric-oxide synthase-2 and the consequent production of nitric oxide is one of the causes of pathogenesis of chronic inflammation in arthritis (McCartney-Francis et al., 2001), with nonspecific inhibitors of nitric-oxide synthase reducing experimental arthritis in animals (McCartney-Francis et al., 1993; Connor et al., 1995). However, selective inhibitors of inducible nitric-oxide synthase have been found to exacerbate the chronic inflammatory response in experimental arthritis (Abramson et al., 2001; McCartney-Francis et al., 2001). Thus, because TNF-α regulates nitric-oxide synthase-2 through nitric oxide production (Amin et al., 1999; Fahmi, 2004) and because cucurbitacin R was shown to reduce both TNF-α production and nitric-oxide synthase induction in cell culture without modifying the enzyme activity, this compound may actually be effective in reducing tissue damage, as the suppression of nitric oxide after treatment with TNF-α antagonists results in a reduction of inflammation and of the associated synovial fluid pathology (Wahl et al., 2003).
Cucurbitacin R may be responsible for modifying nuclear factor-κB activation, because it not only inhibited the nitric oxide generation from murine macrophages activated with lipopolysaccharide and interferon-γ but it also blocked nuclear factor-κB activation, which is essential for the transcriptional activation of nitric-oxide synthase induction (Park et al., 2001). Still, in our most recent experiment, cucurbitacin R did not seem to affect nuclear factor-κB activation (data not shown). The difference in results could be due to the effect on the activation of the Janus kinase-2 and STAT3 pathway, which implicates the induction of apoptosis and inhibition of cell growth via the modification of extracellular transduction from cytokines and growth factors, especially because STAT3 has been shown to inhibit immune responses by blocking the expression of proinflammatory factors (Wang et al., 2004). In fact, Sun et al. (2005) studied some related cucurbitacins and demonstrated that cucurbitacin Q inhibits the activation of STAT3, cucurbitacin A inhibits Janus kinase-2, and cucurbitacins B, E, and I inhibit the activation of both. This produces a consequent decrease in the proinflammatory factors, including TNF-α and interleukin-1β, which are produced by these cells in tissue as demonstrated by the fact that a negative effect on STAT3 reduces TNF-α production in experimental arthritis while increasing the apoptosis of macrophages, synovial fibroblasts, and lymphocytes (Liu and Pope, 2003). Our experimental data confirm the inhibition of STAT3 activation in lymphocytes from both healthy and arthritic men. Moreover, Graness et al. (2006) recently studied cucurbitacin I and demonstrated that it produced alterations in the actin cytoskeleton and focal adhesions, which may modulate cell morphology, migration, adherence, and gene expression.
In conclusion, cucurbitacin R demonstrated an anti-inflammatory effect on an experimental model of induced arthritis, reducing both tissue and bone damage. The mechanism of action involves the reduction of TNF-α production, as well as the production of both nitric oxide and prostaglandin E2. In addition, the induction of nitric-oxide synthase is reduced, but there is no effect on cyclooxygenase. In these features, the inhibition of STAT3 activation is implicated, whereas that of nuclear factor-κB is not.
Acknowledgments
We are indebted to the Centre de Transfusions de la Comunitat Valenciana (Valencia, Spain) for generous supply of human blood and to Dr. Fernando Ribas for radiographic analysis.
Footnotes
-
This work was supported by the Spanish Government (SAF2002-00723), Generalitat Valenciana (GV04B/230), and Universitat de Valencia (UV-AE-06-14). J.M.E. is the recipient of a fellowship from Generalitat Valenciana (Grant CTBPRB/2003/315).
-
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
-
doi:10.1124/jpet.106.107003.
-
ABBREVIATIONS: TNF-α, tumor necrosis factor-α; PBS, phosphate-buffered saline; MTT, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenil-tetrazolium bromide; l-NAME, Nω-nitro-l-arginine methyl ester; STAT, signal transducer and activator of transcription.
- Received April 27, 2006.
- Accepted October 24, 2006.
- The American Society for Pharmacology and Experimental Therapeutics