https://orcid.org/0000-0002-8288-4254
Abstract
To investigate the main chemical components and the anti-inflammatory activity of extracts of Adelia ricinella L. aerial parts.
Three extracts obtained by soxhlet extraction and ethanol/water mixtures were evaluated in their chemical composition by UPLC-DAD-MS/MS. The in vitro anti-inflammatory activity of the prepared extracts was assessed through three different assays: COX-1 and COX-2 enzymatic inhibition, cell-based COX assays on RAW264.7 macrophages (ATCC) measuring the COX-2 protein expression by Western blot and the measurement of the PGE2 concentration in the supernatants of the culture medium. Also was determinate the effect of the three extracts on the RAW 264.7 cell viability.
Few differences in the phytochemical profile were found between the three prepared extracts, identifying a blend of thirteen flavonoids derived from luteolin and apigenin, with orientin as main constituent. Plant extracts (alcoholic and aqueous) did not affect the macrophage cell viability (IC50 > 256 μg/ml) and significantly reduced COX-1 and COX-2 enzyme activities. Additionally, COX-2 expression and PGE2 release were suppressed after 24 h of LPS stimulation and treatment with plant extracts (8–64 µg/ml).
A. ricinella extracts showed the ability to reduce the inflammatory effect exerted by LPS in murine macrophages. However, further studies should confirm their anti-inflammatory activity.
Introduction
Adelia ricinella L. is a tree native from the Caribbean belonging to the family Euphorbiaceae. It has been described in all regions of Cuba, Jamaica and Cayman Islands, as well as in coastal areas of Colombia and Venezuela. The ethnobotanic information refers to it use as medicinal plant, attributing as main activities an antipyretic, analgesic and anti-inflammatory effect.[1] In spite of this strong ethno-pharmacological evidence, from the scientific point of view, only little information about the chemical or pharmacological potential is available. Previous work demonstrated that A. ricinella produces high quantities of polyphenols and flavonoids, being responsible for the in vitro antioxidant activity.[2] Nevertheless, they have not been identified yet.
Macrophages actively participate in the development of inflammation-related pathological processes such as auto-immune diseases, arthritis and cancer.[3, 4] These cells can be activated by various stimuli like bacterial and fungal components and chemical mediators.[5] Afterward, they release pro-inflammatory proteins (IL-1β, IL-6 and TNF-α) and other inflammatory mediators such as nitric oxide and prostaglandin E2 (PGE2), etc. which are responsible for several biological functions, although resulting in damage in case of overproduction of these mediators.[6] One of the main targets to reduce the negative effect of an intensified inflammatory reaction is to address the COX enzyme system. COX-1 and COX-2 mediate the eicosanoids release like PGE2 and thromboxanes, which are well known to participate in chronic inflammation and tumour formation.[7] In this sense, the use of plant-based active products with anti-inflammatory potential, able to decrease COX activities, is considered as an important goal for the development of new biopharmaceuticals lacking adverse effects.[8]
This research was aimed to evaluate the in vitro anti-inflammatory effect of alcoholic and aqueous extracts of Adelia ricinella leaves by measuring COX-1 and COX-2 enzyme suppression, and PGE2 release in lipopolysaccharide (LPS)-activated RAW 264.7 murine macrophages, as well as to determine the main chemical constituents in these extracts. This study may allow a better understanding of the traditional uses of this plant species by the local population against several inflammation-related diseases, establishing for first time a relationship between the pharmacological activity measured and the phytochemical composition.
Materials and Methods
Materials and reagents
Acetonitrile (HPLC grade) (ACN) was acquired from Fisher Chemical UK Ltd. Formic acid (FA) (98+%, analytical grade) was obtained from Acros Organics (Belgium). The external standards isovitexin, orientin and isoorientin (all with purity ≥ 99%) were from Extrasynthese (Lyon, France), while vitexin (99.7%) was purchased from Adipogen (Liestal, Switzerland).
Lipopolysaccharide (LPS) from Escherichia coli (0128:B12), DMSO (Uvasol), porcine hematin, L-epinephrine, Na2EDTA, celecoxib, arachidonic acid, tamoxifen and resazurin sodium salt (7-hydroxy-3H-phenoxazin-3-one-10-oxide) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Tris-buffer was purchase from Biorad. Dulbecco's Medium Eagle Modified (DMEM), Dulbecco's phosphate-buffered saline (DPBS) and 5 ml of ram seminal vesicle COX-1 (1 U) or human recombinant COX-2 (0.5 U) were also from Sigma-Aldrich (St. Louis, MO, USA). Indomethacin was purchased from MP Biomedicals. Na2EDTA (Titriplex III) was purchased from VWR International. Arachidonic acid, purified COX-1 from ram seminal vesicles and human recombinant COX-2 were from Cayman Chemical. The competitive PGE2 EIA kit was purchased from Enzo Life Science
Plant material and extract preparation
The aerial parts of Adelia ricinella L. were collected in the Siboney-Juticí Ecological Reserve (latitude: 19.9603 and longitude: −75.7081), located outside the city of Santiago de Cuba. The plant was also taxonomically identified at different moments by specialists at the Eastern Centre for Ecosystems and Biodiversity (BIOECO, Santiago de Cuba) and a vegetal sample is already settled at the herbarium of the institution with the registration number 14 780.
The extracts were prepared by Soxhlet extraction (VWR, Leuven, Belgium) during 4 h after the first reflux using either water, ethanol 50% and ethanol 95% as described by Berenguer et al. 2018.[2] Afterward, the extracts AR1 (EtOH 95%), AR2 (EtOH 50%) and AR3 (aqueous extract) were filtered using a Büchner funnel and filter paper, and finally evaporated to dryness under reduced pressure in a rotary evaporator (Kika Werke RV 05, Germany) below 45ºC.
Analysis of chemical composition by UPLC-DAD-MS/MS
Chemical composition analysis was performed in an UPLC-DAD-MS/MS system using a Xevo G2-XS QTof spectrometer (Waters, Milford, MA, USA) coupled with an ACQUITY LCsystem equipped with MassLynx version 4.1 software. For analysis, 5 µl of each extract (AR1-AR3) at 100 µg/ml were injected on a BEH Shield RP18 column (100 mm × 2.10 mm, 1.7 µm, Waters, Milford, MA, USA). The mobile phase solvents consisted of H2O + 0.1% FA (A) and ACN + 0.1% FA(B), and the gradient was set as follows (min/B%): 0.0/10, 5.0/10, 20.0/15, 30/15, 40.0/25, 45.0/25, 55.0/40, 60.0/40, 65.0/100, 70.0/100, 75.0/10 and 85.0/10. Full scan data were recorded in ESI (−) and ESI (+) mode from m/z 100 to 1500 and the analyzer was set in sensitivity mode (approximate resolution: 22 000 FWHM). The spray voltage was set at either +1.5 kV and −1.0 kV; cone gas flow and desolvation gas flow at 50.0 L/h and 1000.0 L/h, respectively; and source temperature and desolvation temperature at 120°C and 550°C, respectively. Data were also recorded using MSE in the positive and negative ionization modes (two analyses per mode), and a ramp collision energy from 20 till 30 V was applied to obtain additional structural information. Leucine-encephalin was used as lock mass. DAD spectra were recorded between 190 and 500 nm.
COX-1 and COX-2 enzymatic inhibition assay
COX-1 and COX-2 inhibition assays were performed in a 96-well plate as previously described the literature.[9] The incubation mixture contained 180 μl of 0.1 M Tris HCl-buffer (pH 8.0), 5 μm hematin, 18 mm epinephrine bitartrate, 0.2 U of enzyme preparation and 50 μm Na2EDTA (only for COX-2 assay). Each sample solution (10 μl) was added and the mixture was preincubated for 5 min at room temperature. The extracts were dissolved in DMSO at different concentrations (2–256 µg/ml, DMSO concentration did not exceed 0.1%. Indomethacin (1.25 µm in EtOH abs.) and celecoxib (2.5 µm in DMSO) were used as positive controls. The reaction was started by adding 10 μl of 5 μM arachidonic acid in ethanol p.a. and incubated at 37ºC. After 20 min the reaction was stopped by the addition of 10 μl of 10% formic acid.
Cell-Based COX Assays
Cell culture and reagents
RAW264.7 macrophages from ATCC (American Type Tissue Culture Collection, USA) were maintained at 37°C, 5% CO2 atmosphere in DMEM medium supplemented with 10% FCS, 2% L-glutamine and 4.5 g/L D-glucose.
Macrophage cell suspension was seeded in a six-well plate (5 × 105 cells/ml) and incubated at 37°C in a humidified atmosphere with 5%. After 48 h, the old medium was discarded and the cells were pre-incubated with and without Adelia ricinella extracts (ranging 8–64 µg/ml) and LPS (100 ng/ml) for 24 h.
COX-2 Protein Expression in Macrophages
Western blot analysis
Cells were washed twice with PBS, scraped into 0.5 ml of ice-cold extraction lysis buffer (50 mm Tris pH 8.0, 150 mm NaCl, 1% Nonidet P-40) and maintained for 30 min at 4°C in constant agitation. Cell debris was removed by micro-centrifugation at 15.000×g during 30 min at 4°C and supernatants were rapidly frozen. Protein concentrations in the supernatant were evaluated using the BCA Protein Assay Kit (Termo Fisher, Montluçon, France) as suggested by the manufacturer.
Total proteins (15 µg) were separated by electrophoresis on 12.5% polyacrylamide gels and transferred for 1 h onto nitrocellulose membranes. The membranes were then blocked for 1 h with 5% BSA Tris-buffered saline/0.1% Tween 20 and incubated with primary antibodies against COX-2 (1/1000) from Cell Signaling (Ozyme, Saint Quentin Yvelines, France). Then, membranes were incubated with secondary fluorescent goat anti-rabbit and goat anti-mouse IgG (H + L) antibodies (1/20 000) from Cell Signaling (Ozyme, Saint Quentin Yvelines, France). Proteins were visualized by enhanced chemiluminescence and imaged using a Syngene G:BOXChemi XR system and GeneSnap software (Version 7.08.11; Syngene USA) and the signals was quantified using Image J software. Samples were normalized by GAPDH housekeeping protein.
PGE2 measurement
After collecting the supernatants, the PGE2 concentration in the culture medium was quantified using the EIA kit for PGE2 according to the manufacturer's protocol. The production of PGE2 was measured relative to the control treatment. All experiments were performed in triplicate.
Cell viability assay
The effect Adelia ricinella extracts on the RAW 264.7 cell viability was determined by the Resazurin dye reduction test. Briefly, 200 µl of cell suspension (5 × 105 cells/ml) was added in a 96-well plate and incubated in the same conditions mentioned above. After 24 h, the cells were washed twice with 200 µl of DPBS and then fresh DMEM without FCS was added. The plant extract (8, 16, 32, 64, 128 and 256 µg/ml) were added into each well and incubated at 37°C, 5% CO2. Subsequently, 50 µl of Resazurin solution (2.2 µg/ml) was added and the fluorescence was measured after 4 h (λ excitation 550 nm, λ emission 590 nm) using a TECAN GENios microplate reader (Männedorf, Switzerland). Two independent experiments were performed and the samples were tested in triplicate. Tamoxifen was included as a reference of cytotoxicity.
Statistical analysis
Statistical analysis was performed using the statistical software package GraphPad Prism 7 (Windows, V. 7.04, 2017). All results were statistically analyzed and expressed as the arithmetic means ± standard deviation (SD). One-way ANOVA test followed by the Tukey test was applied to determine the significance of differences between groups. Differences at P ≤ 0.05 were accepted as significant.
Results and Discussion
Phytochemical analysis
Three extracts of Adelia ricinella, that is AR1 (ethanol/water 95:5% v:v), AR2 (ethanol/water 50:50% v:v) and AR3 (water) were analyzed by UPLC-DAD-MS/MS. The total ion chromatogram (TIC) of all extracts (Figure 1) showed that most of the peaks are present in these three extracts, being only different by the intensity of the signal. The alcoholic extracts showed more defined and higher peak areas than the aqueous extract. Thirteen peaks were defined (peaks with an intensity level greater than 30% in any extract and at least one well-defined peak in one of the other two).
UPLC-DAD-MS/MS TIC profiles of Adelia ricinella L. extracts.
Table 1 shows the thirteen compounds identified with it most important signals in the ESI negative ion mode mass spectrum that helped mainly to elucidate the glycosidic type compounds. ESI positive ion mode was useful for determination of the nature of the aglycones. The thirteen peaks defined, indicated by a retention time range represented by the maximal signal appeared in each extract, were assigned as follows: The mass spectrum of peaks 1 (Rt from 9.59 to 9.64 min) and 2 (Rt from 9.77 to 9.79 min) both displayed a pseudo-molecular ion at m/z 609.1464 [M−H]− but with a different fragmentation pattern. While in peak 1 product ions were observed at m/z 447.0935, 327.0883 and 285.0407, peak 2 showed characteristic ions at m/z 461.1330, 327.0714 and 285.0405. In both cases, fragments from the ESI in the positive ion mode were observed at m/z 161.0227 and 179.0347, characteristic for the 0,4B+-H2O and 0,4B+ cleavage of the C-ring of a flavone type aglycon, which was in agreement with luteolin and made it possible to make the distinction between kaempferol which has the same molecular weight. Clearly both compounds were di-glycosides of luteolin, but compound 1 corresponded to a luteolin-C,O-di-hexoside in which the O-hexoside unit was lost first; while compound 2 corresponded to a luteolin-di-C-hexoside, with each hexose linked at a different position, therefore C6 and C8. Compound 3, eluted with Rt from 10.45 to 10.46 min, and the [M−H]− ion was observed at m/z 579.1365. It showed a characteristic m/z 285.0403 fragment ion in the MS2 experiment in negative ion mode, corresponding to [M-162-132-H]− consistent with the loss of an O-linked-hexoside-pentoside disaccharide residue. Once again, in the positive ion mode fragments at m/z 161.0227 and 179.0337 were observed, associated to the luteolin nucleus. Peak 4 (Rt from 10.92 to 10.94 min) showed an [M−H]− ion at m/z 593.1503 and a single product ion at m/z 285.0402, corresponding to [M-162-142-H]; therefore, it was deduced to be a luteolin-type-O-hexoside-deoxyhexoside (same behaviour in ESI positive ion mode MS). Peak 5, based on the larger areas (UV detection at 256/366 nm) in the chromatograms of the alcoholic extracts, emerged with an Rt from 11.08 to 11.09 min as the main constituent of A. ricinella. Compound 5 showed a pseudo-molecular ion at m/z 447.0927 [M−H]− and product ions at m/z 327.0506 and 285.0402. This fragmentation pattern was consistent with a luteolin C-hexoside, which identity was confirmed to be orientin by analyzing the standards orientin and isoorientin, displaying the same Rt and mass spectral data as orientin. The UV spectra of compounds 1–5 are all quite similar with maxima around 228 and 345 nm, being consistent with different luteolin glycosides.
Assigned compounds, [M−H]− and ESI negative fragment ions of the thirteen flavonoids identified in Adelia ricinella extracts
| Peak No. | [M−H]− (m/z) | Retention time range AR1–AR3 (min) | Molecular formula | Other fragments | Identification (Compound) |
|---|---|---|---|---|---|
| 1 | 609.1464 | 9.59–9.64 | C27H30O16 | 447.0935, 327.0869, 285.0396 | Luteolin-C,O-di-hexoside |
| 2 | 609.1464 | 9.77–9.79 | C27H30O16 | 461.1330, 327.0714, 285.0405 | Luteolin-di-C-hexoside |
| 3 | 579.1365 | 10.45–10.46 | C26H28O15 | 285.0403 | Luteolin-O-hexoside-pentoside |
| 4 | 593.1503 | 10.92–10.94 | C27H30O15 | 285.0404 | Luteolin-O-hexoside-deoxyhexoside |
| 5 | 447.0927 | 11.08–11.09 | C21H20O11 | 327.0506, 285.0402 | Orientin |
| 6 | 563.1397 | 11.73–11.75 | C26H28O14 | 463.0877, 300.0275, 269.0451 | Apigenin-O-hexoside-pentoside |
| 7 | 607.1658 | 12.23–12.25 | C28H32O15 | 577.1551, 299.0554, 269.0451 | Diosmin |
| 8 | 431.0984 | 12.52–12.55 | C21H20O10 | 311.0580, 269.0449 | Vitexin |
| 9 | 653.1716 | 13.20–13.23 | C29H34O17 | 607.1671, 507.1183, 283.0614, 268.0375 | (2-hydroxyethoxy)-O-hexoside-deoxyhexoside-4′-methyl-kaempferol |
| 10 | 637.1764 | 14.44–14.46 | C29H34O16 | 591.1714, 283.0607, 268.0371 | O-glucopyranuronosyl- deoxyhexoside-O-methylapigenin |
| 11 | 285.0483 | 16.55–16.56 | C15H10O6 | – | Luteolin |
| 12 | 577.1349 | 17.25–17.27 | C30H26O12 | 431.0980, 329.2325, 269.0451 | Apigenin-C-(O-p-coumaroyl-hexoside) |
| 13 | 269.0454 | 18.22–18.23 | C15H10O5 | – | Apigenin |
| Peak No. | [M−H]− (m/z) | Retention time range AR1–AR3 (min) | Molecular formula | Other fragments | Identification (Compound) |
|---|---|---|---|---|---|
| 1 | 609.1464 | 9.59–9.64 | C27H30O16 | 447.0935, 327.0869, 285.0396 | Luteolin-C,O-di-hexoside |
| 2 | 609.1464 | 9.77–9.79 | C27H30O16 | 461.1330, 327.0714, 285.0405 | Luteolin-di-C-hexoside |
| 3 | 579.1365 | 10.45–10.46 | C26H28O15 | 285.0403 | Luteolin-O-hexoside-pentoside |
| 4 | 593.1503 | 10.92–10.94 | C27H30O15 | 285.0404 | Luteolin-O-hexoside-deoxyhexoside |
| 5 | 447.0927 | 11.08–11.09 | C21H20O11 | 327.0506, 285.0402 | Orientin |
| 6 | 563.1397 | 11.73–11.75 | C26H28O14 | 463.0877, 300.0275, 269.0451 | Apigenin-O-hexoside-pentoside |
| 7 | 607.1658 | 12.23–12.25 | C28H32O15 | 577.1551, 299.0554, 269.0451 | Diosmin |
| 8 | 431.0984 | 12.52–12.55 | C21H20O10 | 311.0580, 269.0449 | Vitexin |
| 9 | 653.1716 | 13.20–13.23 | C29H34O17 | 607.1671, 507.1183, 283.0614, 268.0375 | (2-hydroxyethoxy)-O-hexoside-deoxyhexoside-4′-methyl-kaempferol |
| 10 | 637.1764 | 14.44–14.46 | C29H34O16 | 591.1714, 283.0607, 268.0371 | O-glucopyranuronosyl- deoxyhexoside-O-methylapigenin |
| 11 | 285.0483 | 16.55–16.56 | C15H10O6 | – | Luteolin |
| 12 | 577.1349 | 17.25–17.27 | C30H26O12 | 431.0980, 329.2325, 269.0451 | Apigenin-C-(O-p-coumaroyl-hexoside) |
| 13 | 269.0454 | 18.22–18.23 | C15H10O5 | – | Apigenin |
Assigned compounds, [M−H]− and ESI negative fragment ions of the thirteen flavonoids identified in Adelia ricinella extracts
| Peak No. | [M−H]− (m/z) | Retention time range AR1–AR3 (min) | Molecular formula | Other fragments | Identification (Compound) |
|---|---|---|---|---|---|
| 1 | 609.1464 | 9.59–9.64 | C27H30O16 | 447.0935, 327.0869, 285.0396 | Luteolin-C,O-di-hexoside |
| 2 | 609.1464 | 9.77–9.79 | C27H30O16 | 461.1330, 327.0714, 285.0405 | Luteolin-di-C-hexoside |
| 3 | 579.1365 | 10.45–10.46 | C26H28O15 | 285.0403 | Luteolin-O-hexoside-pentoside |
| 4 | 593.1503 | 10.92–10.94 | C27H30O15 | 285.0404 | Luteolin-O-hexoside-deoxyhexoside |
| 5 | 447.0927 | 11.08–11.09 | C21H20O11 | 327.0506, 285.0402 | Orientin |
| 6 | 563.1397 | 11.73–11.75 | C26H28O14 | 463.0877, 300.0275, 269.0451 | Apigenin-O-hexoside-pentoside |
| 7 | 607.1658 | 12.23–12.25 | C28H32O15 | 577.1551, 299.0554, 269.0451 | Diosmin |
| 8 | 431.0984 | 12.52–12.55 | C21H20O10 | 311.0580, 269.0449 | Vitexin |
| 9 | 653.1716 | 13.20–13.23 | C29H34O17 | 607.1671, 507.1183, 283.0614, 268.0375 | (2-hydroxyethoxy)-O-hexoside-deoxyhexoside-4′-methyl-kaempferol |
| 10 | 637.1764 | 14.44–14.46 | C29H34O16 | 591.1714, 283.0607, 268.0371 | O-glucopyranuronosyl- deoxyhexoside-O-methylapigenin |
| 11 | 285.0483 | 16.55–16.56 | C15H10O6 | – | Luteolin |
| 12 | 577.1349 | 17.25–17.27 | C30H26O12 | 431.0980, 329.2325, 269.0451 | Apigenin-C-(O-p-coumaroyl-hexoside) |
| 13 | 269.0454 | 18.22–18.23 | C15H10O5 | – | Apigenin |
| Peak No. | [M−H]− (m/z) | Retention time range AR1–AR3 (min) | Molecular formula | Other fragments | Identification (Compound) |
|---|---|---|---|---|---|
| 1 | 609.1464 | 9.59–9.64 | C27H30O16 | 447.0935, 327.0869, 285.0396 | Luteolin-C,O-di-hexoside |
| 2 | 609.1464 | 9.77–9.79 | C27H30O16 | 461.1330, 327.0714, 285.0405 | Luteolin-di-C-hexoside |
| 3 | 579.1365 | 10.45–10.46 | C26H28O15 | 285.0403 | Luteolin-O-hexoside-pentoside |
| 4 | 593.1503 | 10.92–10.94 | C27H30O15 | 285.0404 | Luteolin-O-hexoside-deoxyhexoside |
| 5 | 447.0927 | 11.08–11.09 | C21H20O11 | 327.0506, 285.0402 | Orientin |
| 6 | 563.1397 | 11.73–11.75 | C26H28O14 | 463.0877, 300.0275, 269.0451 | Apigenin-O-hexoside-pentoside |
| 7 | 607.1658 | 12.23–12.25 | C28H32O15 | 577.1551, 299.0554, 269.0451 | Diosmin |
| 8 | 431.0984 | 12.52–12.55 | C21H20O10 | 311.0580, 269.0449 | Vitexin |
| 9 | 653.1716 | 13.20–13.23 | C29H34O17 | 607.1671, 507.1183, 283.0614, 268.0375 | (2-hydroxyethoxy)-O-hexoside-deoxyhexoside-4′-methyl-kaempferol |
| 10 | 637.1764 | 14.44–14.46 | C29H34O16 | 591.1714, 283.0607, 268.0371 | O-glucopyranuronosyl- deoxyhexoside-O-methylapigenin |
| 11 | 285.0483 | 16.55–16.56 | C15H10O6 | – | Luteolin |
| 12 | 577.1349 | 17.25–17.27 | C30H26O12 | 431.0980, 329.2325, 269.0451 | Apigenin-C-(O-p-coumaroyl-hexoside) |
| 13 | 269.0454 | 18.22–18.23 | C15H10O5 | – | Apigenin |
Compound 6 (Rt from 11.73 to 11.75 min) was characterized by apseudo-molecular ion at m/z 563.1397 [M−H]−. The MS2 experiment showed an ion at m/z 269.0454 corresponding to [M-162-132-H]−. Fragments observed in ESI positive ion mode MS (1-3A+and 0-2A+) at m/z 153.0119, characteristic of C1—C3 bond cleavage in the C-ring and an ion at m/z 149.0526 from C0—C2 bond cleavage respectively, were in agreement with an apigenin structure. Altogether, this suggested that compound 6 was an apigenin-O-linked hexoside-pentoside disaccharide. Peak 7 (Rt from 12.23 to 12.25 min) presented an [M−H]− ion at m/z 607.1658 and product ions at m/z 577.1551, 299.0554 and 269.0451. The ion at m/z 299.0554 suggested the loss of [M-162-146-H]−, consistent with an O-hexoside-deoxyhexoside disaccharide, while ions at m/z 577.1571 [M-H-30.0087]− and 269.0451 represented the loss of an OCH3 moiety. The same fragmentation pattern was observed in ESI positive ion mode MS, suggesting that compound 7 was the widespread flavonoid diosmin. Compound 8 (the third most abundant compound considering the peak area with UV detection 256/366 nm), assigned to the peak of the same number (Rt from 12.52 to 12.55 min), produced a pseudo-molecular [M−H]− ion at m/z 431.0984 and product ions at m/z 311.0580 and 269.0449. It was confirmed as being vitexin using an analytical standard. Compound 9 (Rt from 13.20 to 13.23 min) showed its [M−H]− ion at m/z 653.1716 and product ions at m/z 607.1671, 507.1183, 283.0614, and 268.0375 in ESI negative ion mode. The loss of 46 mass units can be associated with a 2-hydroxyethoxygroup, releasing the fragment ion at m/z 607.1671, while the fragment 507.1183 looks like the loss of a deoxyhexose [M-H-146]−. The next fragment appeared at m/z 283.0614, representing the additional loss of a162 mass units related to an O-hexoside linkage. Finally, fragment ion at m/z 268.0375 represented the loss of 15 mass units (–CH3) from m/z 283.0614. The fragmentation pattern in ESI positive ion mode MS shows fragments at 163.0703 and 153.0177 in a 2:3 peak intensity proportion, suggesting that compound 9 contains a 3-OH moiety (fragments 1-3A+and 0-2A+, respectively) and in consequence, indicating that the O–CH3 substitution is placed at the 4′ position. Altogether, this allows to define compound 9 as (2-hydroxyethoxy)-O-hexoside-deoxyhexoside-4′-methyl-kaempferol. Compound 10 (Rt from 14.44 to 14.46 min) was characterized by an [M−H]− ion at m/z 637.1764. The MS2 experiment showed ions at m/z 591.1714, 283.0607 and 268.0371. This compound looked quite similar to compound 9 with similar behavior in ESI positive ion mode, but with a deoxy sugar moiety. This allowed defining it as an O-glucopyranuronosyl-deoxyhexoside-O-methylapigenin derivative. Compound 11 (Rt from 16.55 to 16.56 min, being the second most abundant constituent considering the peak area with UV detection at 256/366 nm) produced an [M−H]− ion at m/z 285.0483 without any other product ions with over 5% relative intensity. Fragments in ESI positive ion mode MS appeared at m/z 153.0181, 161.0231 and 179.0337, consistent with C1–C3 and C0–C4 cleavage (fragments 1-3A+, 0-4B+) being defined as luteolin. On the other hand, compound 13 (Rt from 18.22 to 18.23 min) showed an [M−H]− ion at m/z 269.0454 without any other product ionwith over 5% relative intensity. Fragments observed in ESI positive ion mode MS suggested apigenin as compound 13. Both compounds (11 and 13) were confirmed by adding standards in others run. Compound 12 (Rt from 17.25 to 17.27 min) showed a [M−H]− ion at m/z 577.1349 and a product ion at m/z 431.0980 (low relative intensity), which suggested the loss of a coumaroyl group by cleavage at the carbonyl position, losing a charged fragment producing a peak at m/z 146.0369. Fragments at m/z 329.2325 and 269.0449 were assigned to a different pattern of a C-sugar moiety cleavage as is represented in Figure 2. Once again, fragments observed in ESI positive ion mode MS suggested the apigenin aglycone; therefore, the proposed structure is an apigenin-C-(O-p-coumaroyl-hexoside).
Fragmentation pathway of compound 12 (Positions of the C-linked sugar moiety and O-pcoumaroyl substituent are arbitrary).
COX-1 and COX-2 enzymatic inhibition assay
The in vitro effect of the Adelia ricinella extracts on the enzymatic activity of COX-1 and COX-2 (Figure 3A and B) was firstly evaluated using a cell-free experiment. This type of assay may offer prior information about the enzyme selectivity of bioactive compounds present in plant extracts. The ethanol extracts (AR1 and AR2) strongly inhibited COX-1 and COX-2 enzyme activity. Additionally, both extracts were, in a dose-dependent manner, significantly more active (up to ten times more) for COX-2 (IC50 15.9 and 8.4 µg/ml, respectively) than for COX-1 (IC50 52 and 86 µg/ml, respectively). On the contrary, the aqueous extract (AR3) did not exert any significant inhibitory effect on the enzymes.
Inhibitory effect of Adelia ricinella L. extracts on COX-1 and COX-2 activities and PGE2 release.
Both COXs isoforms are well known to be targets for several nonsteroidal anti-inflammatory compounds. COX-1 is widely distributed and constitutively expressed in most tissues where it is involved in homeostatic functions, mainly in the gastrointestinal tract. However, an intensified inhibition of the enzyme activity by anti-inflammatory drugs may provoke several adverse effects on gastric functions. The COX-2 inducible isoform, more predominant at sites of inflammation, appears to play a key role in pathophysiologic conditions like inflammatory disorders, and has driven the therapeutic development of COX-2 inhibitors.[10]
COX-2 protein expression in macrophages
In the case of COX-2 a cell-based assay was carried out. It was observed by Western blot analysis that the ethanol extracts (AR1 and AR2) significantly reduced COX-2 expression in LPS-stimulated RAW 264.7 macrophages at 64 µg/ml (Figure 3C). Additionally, the aqueous extract (AR3) also exhibited an important decrease of COX-2 expression at 64 µg/ml. When comparing the results of the cell-free enzyme inhibition assays with the cell-based COX-2 expression assays, it can be noted that the ethanol extracts (AR1 and AR2) are active in both models, whereas the aqueous extract is only active in the COX-2 expression assay. It can be hypothesized that the aqueous extract is richer in flavonoid glycosides than the ethanolic extracts. Hence, the bulky and polar glycosidic substituents may hinder interaction with the active site of the enzyme.
It seems a contradiction that the most polar extract (AR3), more rich in glycosidic compounds, which are supposed to have a decreased membrane permeability, also showed inhibitory activity in the COX-2 expression assay. However, it mainly concerns C-glycosides, and it has been reported that such compounds, for example orientin and vitexin, showed inhibition of NF-kB and COX-2 expression in vitro, indicating they are at least in part taken up by the cells.[11, 12]
In vitro and in vivo reports confirm that orientin and vitexin, two of the most abundant compounds in Adelia ricinella, contribute to reduction of inflammation through the decrease of the LPS-induced TNF-α/IL-6 release, leukocyte migration and reduction of the expression of pro-inflammatory enzymes (COX-2, iNOS).[13, 14]
On the other hand, in ‘in vivo’ experiments O-glycosidic bonds can be cleaved, allowing absorption of the aglycones and interaction with the target.[15] Additionally, it is well known that flavonoids may also exert anti-inflammatory effects through their antioxidative mechanisms such as scavenging ROS (reactive oxygen species) and RNS (reactive nitrogen species), etc.[7]
PGE2 measurement
Macrophages treated with the A. ricinella extracts were stimulated with LPS (100 ng/ml) for 24 h, and the PGE2 concentration was estimated in the culture medium. As shown in Figure 3D, the ethanol extracts (AR1 and AR2) inhibited the LPS-induced PGE2 release at 64 μg/ml. The results indicated that the inhibitory effects of A. ricinella leaves extracts on LPS-induced PGE2 production might be provoked by the down-regulation of the COX-2 enzyme, blocking the protein expression as previously described.
Additionally, the inhibition of PGE2 biosynthesis by the flavonoid enriched ethanol extracts (AR1 and AR2) could be closely related with the reduction of COX-2 enzyme inhibition evidenced in the previous cell-free experiment. The IC50 of both in vitro assays demonstrated that the inhibitory effect occurred at similar concentrations. It has been reported that flavonoids can inhibit the biosynthesis of prostaglandins, thromboxanes, leukotrienes by inhibition of the enzymes COX, PLA2 or LOX.[6, 7] The absence of an additive or synergistic effect between inhibition of COX-2 expression and enzyme inhibition, may be related to the observation that, if the COX-2 enzyme is inhibited, the cell may paradoxically counterbalance with more protein expression.[16]
On the other hand, the aqueous extract did not show a statistically significant inhibition on the LPS-induced PGE2 production, despite a slight decrease observed at the highest concentration (64 μg/ml), which may due to the fact that the residual quantity of expressed COX-2 protein is sufficient to maintain PGE2 production.
The extracts were not cytotoxic to RAW 264.7 macrophages cells (IC50 > 256 μg/ml) (Figure 4) following the results obtained from the cell viability assay by Resazurin dye reduction test.
Effect of Adelia ricinella (L.) extracts on RAW 264.7 murine macrophages cell viability (All values are expressed as the arithmetic mean ± SD of 6 replicates).
Conclusion
The current study provided a new understanding regarding the main active constituents and the pharmacological potential of Adelia ricinella crude extracts. The extracts were mainly composed of glycosides of luteolin and apigenin. The C-glycosides orientin and vitexin, and the aglycone luteolin, are the three most abundant compounds in this plant species. The plant extracts demonstrated an inhibitory effect on COX-1 and COX-2 enzyme activities and COX-2 protein expression. The results also suggest that ethanol extracts mainly act on PGE2 release probably by COX-2 enzymatic inhibition. This effect is in agreement with the biological properties attributed to the detected flavonoids. The evidence obtained in this work support in part the ethnopharmacological usage of Adelia ricinella leaves as traditional remedy to treat inflammatory disorders.
Funding
This work was supported by the Belgian Development Cooperation through VLIR-UOS in the context of the Institutional University Cooperation Program with the University of Oriente and the VLIR-TEAM project between Antwerp University and Camaguey and Oriente Universities in Cuba.
Author Contributions
The experiments were conceived and designed by C.A. Berenguer-Rivaz, J.C. Escalona-Arranz and G. Llaurado-Maury. The experiments were performed and the data analysed by C.A. Berenguer-Rivaz, J.C. Escalona-Arranz, A. Van der Auwera, S. Piazza, D. Méndez-Rodriguez and K. Foubert. The experiments were supervised by J.C. Escalona-Arranz, P. Cos and L. Pieters.
Conflict of Interest
The authors declare that there are no conflicts of interests
Data Availability
Original data are available from the author of correspondence.
References
