Cnicin from Centaurea benedicta L. is an active compound against skin in ammation in a mouse model

Centaurea benedicta L., commonly known as “cardo santo,” is used as a tonic, antidepressant, anti-inflammatory, antibacterial, and antiseptic in traditional medicine. This study evaluated the topical anti-inflammatory potential of an extract (ECB) and cnicin (CNI) from C. benedicta leaves in a mouse model. Activity was assessed using the ear edema method with croton oil, phenol, capsaicin, and histamine as phlogistic agents. Myeloperoxidase (MPO), N-acetyl-β-D-glucosaminidase (NAG), nitric oxide (NO), tumor necrosis factor α (TNF-α), interleukin 6 (IL-6), and histopathology were assessed as markers of edema/inflammation. Interaction profiles between CNI and cyclooxygenase-1 and -2, induced nitric oxide synthase, and glucocorticoid receptor were examined with molecular docking. Twenty-four h after induction of inflammation, ECB and CNI treatments decreased the thickness and weight of ears by 39.59%– 94.72%. MPO, NAG, NO, TNF-α, and IL-6 levels were also reduced. Histopathological, treatments reduced edema thickness, leukocytes, and vasodilation. Inflammation induced by phenol and histamine was inhibited by ECB and CNI, and ECB suppressed capsaicin-induced inflammation. CNI interacts with cyclooxygenase-1 and nitric oxide synthase through conventional hydrogen bonds, indicating inhibition of these enzymes. ECB and its compound cnicin reduce chemically-induced inflammation in mice suggesting new possibilities for the treatment of diseases associated with dermal inflammatory processes.

Chemical structure of cnicin.
Rinsed leaves of C. benedicta L. (ECB) were extracted with dichloromethane:ethanol (9:1 v/v). Isolation and identification of CNI followed previously described methods (Queiroz et al. 2021). The purity of cnicin was estimated to be greater than 95% by HPLC-DAD analysis (Queiroz et al. 2021). All phytochemical procedures were carried out at the Nucleus for Identification and Research of Natural Active Principles -NIPPAN, Faculty of Pharmacy, Federal University of Juiz de Fora. For this study, ECB and CNI were provided by Dr. Ademar Alves da Silva Filho, coordinator of NIPPAN.

Animals
Male Swiss mice, 30-35 days old, and weighing 25-30 g, were supplied by the Central Animal Facility/UFJF. Experiments were carried out at the Biomedicinal Chemistry and Applied Pharmacology Laboratory of FF/UFJF. Animals were housed in plastic cages (47  34  18 cm) at temperatures between 22°C-24°C, in a 12/12 h light cycle, with food and water ad libitum. Animals were acclimated for 24 h before use. Experiments followed guidelines of the Brazilian College of Animal Experimentation (COBEA) after approval by the Ethics Committee on Animal Use/UFJF (protocol number 022/2018).
Croton oil-induced mouse ear edema As recommended by Colorado et al. (1991), with minor modifications, ear edema was induced with croton oil (2.5% v/v in 20 µL of acetone) applied on the inner surface of the right ear, while the left ear received 20 µL of acetone. After 15 min, the right ear was treated with 20 µL of ECB (0.1, 0.5, and 1.0 mg/ear), CNI (0.1, 0.5, and 1.0 mg/ear), dexamethasone (0.1 mg/ear) and saline (negative control group). A basal group (non-inflamed and untreated) and another treated with vehicle alone were also included. A digital micrometer (Digital Micrometer IP400 to 25 mm  0.001 mm Digimess 110.284-NEW) was used to measure the thickness of the ears (µm) after six and 24 h to assess the development of edema. Weight was quantified 24 h after treatment. An anesthetic overdose [ketamine (300 mg/kg) and Xylazine (30 mg/kg)] was used for euthanasia. Ear fragments (right and left) were removed using a metallic punch (Richter, 0.6 mm) and weighed on an analytical balance (AY220, Shimadzu ® ). The difference in thickness (µm) and weight (mg) between the right and left ears was taken as a measure of edema. Weighed ear fragments were used for assessing inflammatory markers and for histopathology.

Supernatant preparation
The method of De Young et al. (1989), with minor modification, was used to prepare tissues for analysis. Three ear fragments (6 mm) were crushed in a porcelain gral in an ice bath with sodium phosphate buffer (1 mL, 80 mM, pH 5.4) containing hexadecyltrimethylammonium bromide (0.5%). This material was transferred to a test tube and disrupted with an Ultra Cleaner 1600 ultrasound (Unique ® ) for 10 min. Samples were then centrifuged at 3000 rpm for 10 min and supernatants collected and used for the analysis of total protein, myeloperoxidase (MPO), N-acetyl-β-D-glucosaminidase (NAG), nitric oxide (NO), tumor necrosis factor α (TNF-α), and interleukin 6 (IL-6).

Total protein dosage
The method recommended by Lowry (1951) and modified by Sargent (1987) was used to measure total protein concentrations. Forty µL of supernatant, 360 µL of distilled water and 2 mL of reaction mixture [1.960 mL of alkaline solution (1 g of NaOH plus 5 g of Na2CO3 in 250 mL of water, 20 µL of copper tartrate and 20 µL of copper sulfate)] were mixed in test tubes. After 10 min, 200 µL of Folin solution (1:5) was added and tubes were allowed to sit for 30 min. Absorbance at 660 nm was then measured using a Shimadzu ® UV-VIS 1800 spectrophotometer). Distilled water was used as a blank. A standard curve of albumin 1.0 mg/mL (80-400 µg/mL) was also established.

Myeloperoxidase assay
Assays used of the method from Bradley et al. (1982), modified by De Young et al. (1989). Microplates (96-well, n = 3) with 70 µL of supernatant, 35 µL of 3.3', 5.5'-tetramethylbenzidine (1.6 mM in dimethylsulfoxide) and 105 µL 0.003% hydrogen peroxide (v/v, diluted in 80 mM sodium phosphate buffer, pH 5.4) were incubated at 37°C for 5 min. One hundred and forty µL of 4M sulfuric acid at 4°C was added to each well to stop the reaction. Absorbance was measured at 450 nm in a microplate reader, using distilled water as blank. Results are expressed as optical density/mg of protein (OD/mg of protein).

N-Acetyl-β-D-glucosaminidase assay
Using the method of Sánchez and Moreno (1999), triplicate 200 µL aliquots of supernatant were added to 96-well plates along with 50 µL 2.24 mM p-nitrophenyl-N-acetyl-β-D-glucosamine in citrate buffer/100 mM sodium phosphate (pH 4.5). Plates were incubated at 37°C for 10 min and the reaction was stopped with 100 µL 200 mM glycine buffer (pH 10.6). Absorbance at 405 nm was recorded with a microplate reader and results expressed as optical density/mg of protein (OD/mg of protein).

Nitric oxide assay
Nitric oxide (NO) levels were analyzed indirectly by measuring nitrite concentrations (NO 2− ) using the Griess colorimetric method (Green et al. 1982). Griess reagent consists of a mixture (1:1) of sulfanilamide (1%, w/v) and α-naphthyl-ethylenediamine (0.1%, w/v) in 5% phosphoric acid. Homogenates were prepared from ear fragments with 3000 µL of buffered saline (PBS -pH 7.2) and crushed for about 60 s. The homogenate was collected and stored in microtubes at −80°C. The assay was performed in triplicate using a microplate with homogenate (100 µL) and Griess reagent (100 µL). The microplate was incubated at room temperature for 20 min. Absorbance at 540 nm was measured with a microplate reader. Nitrite concentrations were calculated from a standard curve using sodium nitrite solution (NaNO2) (3.12-200 µM). Results are expressed as µMs.

Cytokine assays
TNF-α and IL-6 levels were assessed using the homogenate obtained from ear fragments after centrifugation. Kits that use monoclonal antibodies for pro-inflammatory cytokines, TNF-α and IL-6, were used. Standards with known concentrations were employed following manufacturer's instructions (Peprotech, Rocky Hill, New Jersey, USA). Sensitivities of kits were TNF-α (10-2500 pg/mL) and IL-6 (3-3400 pg/mL). Absorbance at 405 nm was measured with a microplate reader. Cytokine levels were estimated by interpolation from standard curves according to the manufacturer's instructions. Values are expressed in pg/mL.

Histopathological analysis
Ear fragments (6 mm discs) were placed in 10% formaldehyde (v/v) and fixed in 70% ethanol for 24 h (Chibli et al. 2014). Fragments were then dehydrated, blocked in paraffin and transverse sections cut with a microtome (5 μm) (TBS Cut 4060 Rotary Microtome, Thermo Fisher Scientific Inc.). Sections were stained with hematoxylin-eosin and fixed on glass slides for microscopic analysis (BX 51 Olympus microscopy, Olympus Optical Co., LTD; magnification: 20). Representative areas of tissue were selected and Image-Pro ® Plus software (version 6.0, Media Cybernetics, Inc.) used to capture images.
Arachidonic acid-and phenol-induced mice ear edema Ear edema was induced in mice (n = 8) by topical administration on the inner surface of the right ear using arachidonic acid (2.0 mg in 20 µL of acetone/ear) (Young et al. 1984) and phenol (10% v/v in 20 µL of acetone/ear) (Lim et al. 2004), while the left ear received 20 µL of vehicle acetone. Fifteen (arachidonic acid and phenol) after induction of edema, the right ear was topically treated with ECB and CNI (0.1, 0.5 and 1.0 mg/ear in 20 µL of acetone), indomethacin (2.0 mg/ear in 20 mL of acetone, positive control for AA), dexamethasone (0.1 mg/ear 20 µL of acetone, positive control for phenol) and acetone (20 mL/ear, negative control). The animals were euthanized with ketamine (300 mg/kg, i.p.) and xylazine (30 mg/kg, i.p.) 1h after AA and 2h after phenol application for the removal of ear fragments (6 mm discs) using a metal punch (Richter, São Paulo, SP, Brazil). After euthanasia, the ear edema was evaluated by means of increase in ear thickness (µm) and weight (mg).
Capsaicin-induced mouse ear edema As described by Silva et al. (2018), with minor modifications, right ears of eight mice per group (n = 8) were treated with ECB (0.1, 0.5 and 1.0 mg/ear), CNI (0.1, 0.5 and 1.0 mg/ear), dexamethasone (0.1 mg/ear) and 20 µL of saline (negative control group). After 1 h, edema was induced with 20 µL of capsaicin (0.01 mg/μL, v/v, diluted in acetone) was applied to the right ear; the left ear received 20 µL of acetone. Ears from basal and vehicle-only control mice were again included. After 30 min, animals were euthanized with an anesthetic overdose. An external digital micrometer was used to measure thickness (mm), and weights of ear fragments were determined using an analytical balance (AY220, Shimadzu ® ). Measurements of edema by thickness and weight were calculated by differences between right and left ears.
Histamine-induced mouse ear edema Using the method described by Brand et al. (2002), with modifications, mice were divided into groups of eight animals and treated with ECB (0.1, 0.5 and 1.0 mg/ear), CNI (0.1, 0.5 and 1.0 mg/ear), dexamethasone (0 and 1.0 mg/ear), dexchlorpheniramine (0.1 mg/ear) and saline (2 µL, negative control group). Basal and vehicle-only groups were included. Thirty minutes after treatment, animals were anesthetized (100 mg/kg ketamine and 10 mg/kg xylazine) and 10 µL of histamine (100 µg/mL) in saline was applied intradermally in the right ear using a 30G hypodermic needle. In the left ear, the same volume of saline was administered. The digital micrometer was used to measure ear thickness (mm) every 15 min up to 90 min. After 90 min, animals were euthanized with an anesthetic overdose and ear fragments were removed and edema assessed using differences between the right and left ears.

Molecular modeling
Two-dimensional structures of ligands were drawn in MarvinSketch (16.7.4), and threedimensional structures generated using the Clean in 3D function. Structural geometry was refined by semiempirical calculations with MOPAC 2012 applying the PM7 method. 3D crystallographic coordinates (COX-1, code 1EQG; COX-2, code 5IKV; i-NOS, code 1R35; and glucocorticoid receptor, code 1P93) were obtained from the Protein Data Bank (Stewart 2014). Molecular docking used AutoDockVina 1.1.2 simulation boxes (Trott and Olson 2010). Docking calculations were obtained and crystallographic ligands validated by redocking. Discovery Studio v. 4.5 2016 analyzed interactions for molecular recognition.

Statistical analysis
Results are presented as means ± S.E.M. Analysis of variance (ANOVA) followed by post-hoc Student-Newman-Keuls tests were used to assess significance (p < 0.05) using GraphPad Prism ® 7.0 program.

Histopathology
Ear fragments obtained after induction of inflammation with croton oil were analyzed histopathologically to confirm findings from thickness and weight, myeloperoxidase, N-acetyl-β-Dglucosaminidase, and nitric oxide analyses. Representative sections from ears of negative control mice showed marked increases in thickness characteristic of edema induced by typical inflammatory processes. Edema was accompanied by significant presence of leukocytes and vasodilation (Fig. 3A). Treatment with ECB (1.0 mg/ear, Fig. 3C) and CNI (1.0 mg/ear, Fig. 3D) reduced ear thickness and suppressed leukocytes and vasodilation. The positive control, Dexa (0.1 mg/ear), was also effective (Fig. 3B). Effects of ECB and CNI on AA, phenol, and capsaicin-induced mouse ear edema Ear edema/inflammation induced by AA increased the thickness and weight of ears from negative control mice (Fig. 4A and 4B). ECB (23.30, 34.01, and 42.21%) and CNI (16.72,25.99,and 37.07%) reduced edema thickness in a dose-dependent manner (Fig. 4A). This effect was reproduced in the assessment of ear weight, where ECB (40.75,48.83,and 66.79%) and CNI (17.41,27.07, and 57.34%) produced significant results (p < 0.001) (Fig. 4B). IND, as a reference, significantly reduced ear thickness and weight (p < 0.001). Further, effects of ECB (0.1 mg/ear) are not significantly different (p < 0.05) from CNI (0.5 mg/ear) (Fig. 4A).
Induction of inflammation with capsaicin produced a significant increase in thickness and weight of ears of negative control mice 30 min after dermal application ( Fig. 4E and 4F). ECB inhibited this increase in thickness by 37.75, 43.11, and 58.10% at doses of 0.1, 0.5 and 1.0 mg/ear, respectively. ECB also inhibited weight gain by 34.01 (0.1 mg/ear), 44.24 (0.5 mg/ear) and 55.80% (1.0 mg/ear). However, CNI was not effective in reducing capsaicin-induced inflammatory processes. Conversely, CNI (1.0 mg/ear) potentiated the effect of capsaicin on weight gain. Dexa (positive control) was active in inhibiting the thickness and weight of edema ( Fig. 4E and 4F). ECB (1.0 mg/ear) produced an effect similar to that of Dexa (p < 0.05) in reducing thickness (Fig. 4E). All doses of CNI showed similar activity as Dexa (Fig.  4F).

Effects of ECB and CNI on histamine-induced mouse ear edema
Fifteen-minute ECB pretreatment (0.1, 0.5, and 1.0 mg/ear) significantly reduced ear thickness 2 h after intradermal administration of histamine; this effect was dose-dependent (Fig. 5). CNI pretreatment (1.0 mg/ear) inhibited histamine-induced edema 30 min after intradermal administration. This effect was also observed 90 min after injection for CNI (0.1 and 0.5 mg/ear) (p < 0.05; p < 0.001). However, 0.1 mg/ear of CNI potentiated the effect of histamine at 60 and 75 min following histamine administration. Dexchlorpheniramine was more potent than Dexa in reducing edema, though both drugs were effective at all times evaluated (Fig. 5).

Discussion
Lack of effectiveness of drug treatment for some skin diseases encourages the search for new therapeutic agents with fewer adverse effects. This search remains a challenge, especially for natural products (Hahnel et al. 2017). The focus of this study was evaluation of CNI and ECB from C benedicta as topical anti-inflammatory agents using croton oil, Arachidonic acid (AA), phenol, capsaicin, and histamine as inducers of skin inflammation in mice. These agents are advantageous since they are associated with different inflammatory mechanisms and cause pathological conditions that mimic human and animal disease (Patil et al. 2019).
Due to the presence of TPA (12-O-tetradecanoylphorbol-13-acetate), the croton oil induces increased vascular permeability, vasodilation via histamine release, and migration of polymorphonuclear cells and macrophages with skin inflammation and cellular hyperproliferation that mimic skin pathologies, such as psoriasis (De Young et al. 1989). In addition, TPA upregulates gene expression of COX-2, lipoxygenase (LOX), and protein kinase C (PKC), and modulates several pro-inflammatory enzymes (COX-2 and i-NOS), pro-inflammatory proteins (IL-1, IL-2, IL-6, IL-8, TNF-α), and adhesion molecules (Pungeró et al. 1998;Murakawa et al. 2006). Our results showed that, after topical application, ECB and CNI significantly reduced thickness and weight of mouse ears after induction of edema with croton oil (Figures 2A and 2B). CNI (0.5 and 1.0 mg/ear) reduced ear thickness to an extent similar to Dexa six hours after treatment ( Fig. 2A and 2B). Conversely, after 24 hours, the effects of ECB (0.5 and 1.0 mg/ear) were potentiated or equal to Dexa (Fig. 2B and 2C). At this time, CNI was more potent than Dexa at all doses. So, as noted, 24 hours after treatment, variation in thickness and weight were dose-dependent for both agents, demonstrating that a single dose promotes less variation in the anti-inflammatory response. These agents can have mechanisms of action similar as glucocorticoids on nuclear receptors, since they may stimulate or inhibit protein synthesis (Ramamoorthy and Cidlowski 2016). Thus, various cellular functions are thus regulated, including actions of enzymes and synthesis of autacoids and cytokines (Ramamoorthy and Cidlowski 2016). Still, the anti-inflammatory activity of ECB and CNI may also be related to inhibition of other targets, which may provide additional benefits to patients with difficult-to-treat skin disorders, such as psoriasis and related pathologies (Chibli et al. 2014).
Vasodilation and leukocyte migration induced by croton oil is reflected in increased MPO activity that indicates the presence of polymorphonuclear cells (Bradley et al. 1982). A reduction in MPO values in tissues treated with ECB or CNI indicates inhibition of polymorphonucleated leukocytes (neutrophils) infiltration and suppression of inflammation. Also, ECB was as efficacious as Dexa, and CNI was even more potent (p < 0.001) (Fig. 2D), consistent with data presented in Fig. 2A, 2B, and 2C. Infiltration of mononuclear cells was indirectly quantified by assessing NAG activity (Mendes et al. 2009). ECB and CNI reduced the activity of this enzyme (Fig. 2E), indicating a decrease in mononuclear cells and vasodilation. CNI was more potent than Dexa (Fig. 2E), consistent with data presented in Fig. 2A, 2B, and 2C. Also, the release of NAG, a lysosomal enzyme involved in the generation of mediators, such as histamine, serotonin, cytokines, chemokines, and eicosanoids, is characteristic of the accumulation of mononuclear cells in ear edema induced by croton oil (Sánchez and Moreno 1999;Gábor 2003).
Nitric oxide (NO) is a small, simple molecule produced by the action of NO synthase (NOS). This enzyme displays two main types of isoforms, constitutive (c-NOS) and inducible (i-NOS) (Sobrevia et al. 2016). NO produced in endothelial cells diffuses rapidly into muscle cells and vascular lumens and promotes vasodilation. The chemical is involved in both physiological and pathophysiological processes (Mutchler and Straub 2015;Florentino et al. 2017). ECB and CNI reduced concentrations of NO in inflamed tissue (Fig. 2F), corroborating findings for MPO and NAG levels ( Fig. 2D and 2E). Both ECB and CNI were more potent than dexamethasone in inhibiting the formation of NO, which may reflect inhibition of NOS (Fig. 2F). Again, these results are consistent with data for thickness, weight, MPO, and NAG parameters. All findings are consistent with the significant anti-inflammatory activity of ECB and CNI.
TNF-α is released at sites of inflammation and promotes activation of several cell types that release new cytokines and chemical mediators that expand the inflammatory process (Shinwan et al. 2019). Interactions with its receptors, TNFR1 or TNFR2, may activate nuclear transcription factor kappa B (NF-kB) and activating protein-1 (AP-1), as well as TNF-α can induce apoptosis, the formation of reactive oxygen species, and cell necrosis (Yang et al. 2018). In addition, IL-6 is a pro-inflammatory cytokine related to acute amplification of inflammation (Yang et al. 2018). ECB and CNI reduced levels of TNF-α and IL-6 ( Fig. 2G and 2H), consistent with thickness and weight data and MPO, NAG, and NO findings ( Fig. 2D-2F). These data also agree with findings of Schneider and Lachner (1987) that describe antiinflammatory effects of CNI from C. benedicta on rat paw edema.
Pronounced edema, vasodilation, and a greater number of leukocytes were observed in the ears of negative control mice (Fig. 3). Both ECB and CNI also reduced edema thickness, vasodilation, and leukocyte infiltration, consistent with physical and biochemical data for thickness and weight, MPO, NAG, NO, TNF-α and IL-6 ( Fig. 3). ECB and CNI may act by mechanisms similar to Dexa, since is the latter agent is widely used as an adrenocorticoid, antiasthmatic, antiallergic, and anti-inflammatory drug for the treatment of dermatological disorders (Mehta et al. 2016).
AA was applied topically to stimulate prostaglandin production and, consequently, to induce ear edema with the formation of erythema, vasodilation, and leukocyte migration (Young et al. 1984;Doherty et al. 1988). ECB and CNI suppressed ear edema (thickness and weight) in a dose-dependent manner. However, ECB and CNI may also be associated with COX inhibition, since indomethacin, a non-selective COX inhibitor, inhibited the inflammatory process. Further, ECB showed similar effects to CNI, thus corroborating the croton oil-induced ear edema model.
Phenol causes changes in the skin that involve rupture of keratinocytes and release of IL-1α, IL-8, TNF-α, AA, and free radicals. It produces effects that mimic contact dermatitis in humans and animals (Lim et al. 2004;Murray et al. 2007). ECB and CNI counter the effects of phenol (Fig. 4), probably by inhibiting the generation of AA metabolites and/or free radicals. This finding is consistent with data from croton oil and AA models. Moreover, similar effects of ECB and CNI may reflect the cnicin content of ECB, although other ECB constituents may also be involved. These findings show that ECB and CNI might be effective against contact dermatitis.
Capsaicin promotes neurogenic inflammation through activation of transient receptor potential vanilloid 1 (TRPV1), a calcium-permeable channel. This activation promotes the release of neuropepitides (substance P, peptides related to the calcitonin gene and tachykinins) and monoamines (serotonin and histamine) (Gábor and Rázga 1992;Inoue et al. 1993). As a consequence, signs of acute inflammation, such as vasodilation with increased blood flow and elevated local temperature, are triggered (Szolcsányi 1988). The topical application of ECB significantly reduced ear thickness and weight ( Fig. 5A and 5B). Conversely, CNI was ineffective. A dose of 1.0 mg/ear actually potentiated the effect of capsaicin, as indicated by increased ear weight (Fig. 5B). Therefore, CNI does not block the action of capsaicin in activating TRPV1.
Histamine is a vasoactive amine released by activated mast cells. The amine promotes immediate hypersensitivity reactions related to the pathogenesis of various allergic diseases, such as atopic dermatitis, allergic rhinitis, and allergic asthma (Brand et al. 2002). These actions are mediated by HR1 receptors (Thangam et al. 2018) that promotes differential regulation of Th2 lymphocytes with cytokine secretion (IL-5, IL-4, IL-10, and IL-13) and Th1 with inhibition of γ-interferon (IFN-γ), IL-12, and IL-2. Histamine also induces the release of leukotrienes, cytokines, and chemokines (Jemima et al. 2014). ECB reduced ear thickness dose-dependently 15-90 min after histamine injection (Fig. 6). Interestingly, at 60 and 75 min after injection of histamine, CNI potentiated histamine action ( Fig. 6D and 6E). Receptor ligands can affect the stability of active (R*) and inactive (R) states and function either agonistically or antagonistically (Thangam et al. 2018). A similar response was seen when assessing ear fragment weight. ECB produced a similar effect to dexchlorpheniramine. ECB constituents may act as an RH1 antagonist or inverse agonist. Drugs, such as glucocorticoids, cyclosporine, and cromoglycate, show inhibitory effects on mast cell degranulation and mediator release (Amin 2012).
The carboxylic acid group of AA interacts with the active site of COX-1 at Arg120 and Tyr355 residues through hydrogen bonding to promote the production of prostaglandins and other mediators Also, carbon 13 (double-bonded) of AA is close to the phenolic oxygen of Tyr385 with an orientation for H bonding (Marnett et al. 1999). Ibuprofen, a COX-1 inhibitor, hydrogen bonds with the Arg120 residue. CNI also bonds with COX-1 at Arg120 (OH in C15) and Ser530 (OH in C4') residues and via C-H interactions at Tyr355 and Tyr385; the agent may act as a COX-1 inhibitor. The binding affinity of CNI (−7.6 kcal.mol −1 ) for COX-1 is equal to the affinity of the reference compound, ibuprofen (−7.6 kcal.mol −1 ) ( Table 1). Competition for the COX-1 active site might explain the inhibition of edema by CNI after induction by croton oil ( Fig. 2A to 2C) and arachidonic acid ( Fig. 4A and 4B). Such competition would inhibit prostaglandin production. Further, molecular docking indicated that CNI could inhibit inducible nitric oxide synthase (i-NOS) (Sobrevia et al. 2016). Inhibition might occur via hydrogen bonding with Gln192, Tyr276, and Tyr302 residues. The reference compound, I38 interacts in Gln192, Tyr 276, Gly300, Tyr302, and Glu306 by hydrogen bonding. Thus, inhibition of i-NOS could explain the reduction of NO levels shown in Fig. 2F.
Finally, topical anti-inflammatory effects of C. benedicta, may also involve the action of one or several other constituents present in the extract, such as polyphenols (Djamila et al. 2013). Phytochemical constituents, such as tannins and flavonoids, are recognized for their anti-inflammatory effects on wounds. Among the active compounds ECB, CNI is a primary constituent responsible for healing actions (Kataria 1995).

Conclusion
The anti-inflammatory properties of ECB (C. benedicta extract) and CNI (a sesquiterpene lactone) were demonstrated in experimental models of edema/inflammation. Mechanisms of action of these products are related to inhibition of one or more inflammatory response pathways. As an active compound, CNI is a promising anti-inflammatory agent for the treatment of skin diseases, such as psoriasis, contact dermatitis, and atopic dermatitis. The use of these agents may aid the development of new pharmaceutical and dermocosmetic products.