Fatty ethanolamide of Bertholletia excelsa triglycerides (Brazil nuts): anti-inflammatory action and acute toxicity evaluation in Zebrafish (Danio rerio)

Fatty amides (N-alkylamides) are bioactive lipids that are widely distributed in microorganisms, animals, and plants. The low yield in the extraction process of spilantol, a fatty amide, which is mainly related to its diverse biological effects, compromises its application on a large scale. Thus, this study proposes an alternative method to synthesise fatty amides from Bertholletia excelsa (AGBe) oil, with a chemical structure similar to that of spilantol. Carrageenan-induced abdominal oedema in vivo models were used in zebrafish (Danio rerio). In in vivo studies, oral AGBe produced no signs of toxicity. In the histopathological study, AGBe did not cause significant changes in the main metabolising organs (liver, kidneys, and intestines). All doses of AGBe (100 mg/kg, 500 mg/kg, and 750 mg/kg) were effective in reducing oedema by 65%, 69%, and 95%, respectively, producing a dose–response effect compared to the control group, and spilantol-inhibited oedema by 48%. In the in silico study, with the use of molecular docking, it was observed that among the AGBe, the molecules 18:1, ω-7-ethanolamine, and 18:1, ω-9-ethanolamine stood out, with 21 interactions for COX-2 and 20 interactions for PLA2, respectively, surpassing the spilantol standard with 15 interactions for COX-2 and PLA2. The anti-inflammatory action hypothesis was confirmed in the in silico study, demonstrating the involvement of AGBe in the process of inhibiting the enzymes COX-2 and PLA2. Therefore, based on all the results obtained and the fact that until the dose of 1000 mg/kg was administered orally in zebrafish, it was not possible to determine the LD50; it can be said that AGBe is effective and safe for anti-inflammatory activity.


Introduction
Ethnopharmacological studies are increasingly being conducted owing to the growing global demand for new drugs, as well as the presence of chemical components in medicinal plants, which have been associated with several pharmacological effects. Therefore, natural resources have been considered important in improving health and quality of life (Shawahna and Jaradat 2017).
Among the biologically active components reported and found in medicinal plants, we highlight the fatty amides found in abundance in Bertholletia excelsa, a tree of the Lecythidaceae family, also known as chestnut from Pará or Brazil. It is a species native to the Amazon region and is considered one of the main riches of the Amazon jungle. It is the most exported raw material in the region. The oil extracted from the seed of Bertholletia excelsa is notable for its high nutritional value and several biological activities such as healing, antioxidant, and anti-inflammatory activities (Chunhieng et al. 2008;Pena Muniz et al. 2015).
In this context, the species Acmella oleracea, a plant belonging to the Asteraceae family and found mainly in the northern region of Brazil where it is usually used in local cuisine and is popularly known as Jambú, was discovered (Dos . The active compound found in greater abundance in this plant is spilantol (N-alkylamide), a fatty amide with the chemical formula (C 14 H 23 NO,221.339 g/ mol) (Molina-Torres et al. 1996; Barbosa et al. 2015). Spilantol has been used in scientific studies to demonstrate its relationship with several biological effects, such as analgesic, neuroprotective, anticonvulsant, antioxidant, and anti-inflammatory effects (Wu et al. 2008;Hernández et al. 2009;Dias et al. 2011;Silva and Oliveria 2013).
Among the various symptoms, inflammation is the most telling as it is a warning sign for the body, with prolongation of the inflammatory process causing damage to cells and tissues. However, most commercialised anti-inflammatory drugs cause adverse reactions when used in the long term. This has been one of the biggest challenges for doctors and pharmacists who develop research with products of natural origin (Aracama et al. 2000;Shawahna and Jaradat 2017). Considering ethnopharmacology and ethnomedicine, some products of natural origin have been studied and identified as possible alternative drugs for the treatment of inflammation.
In this study, zebrafish (Danio rerio) was selected as an animal model, as it has been recognised pharmacologically for its advantages in carrying out scientific research and validating new drugs, as well as its successful application in the pharmaceutical field (Hsu et al. 2007;Chakraborty et al. 2010;Kettleborough et al. 2013;Schmidt et al. 2013;MacRae and Peterson 2015;Brugman et al. 2016;Carvalho et al. 2017;Borges et al. 2018).
The present study proposed obtaining fatty amides from Bertholletia excelsa oil and have its pharmaco-toxicological validation for anti-inflammatory activity in zebrafish (Danio rerio) in comparison with spilantol, a fatty starch extracted from the flowers of Acmella oleracea, which proved to be an effective anti-inflammatory compound.

Plant materials-obtaining spilantol
Spilantol was previously obtained in a study by de Souza et al. (2020) from the plant species Acmella oleracea, and in this study, it was used as an anti-inflammatory standard for the fatty amides in Bertholletia excelsa oil.
The amidation reaction ( Fig. 1) was carried out with ethanolamine (6.0 mL), BNO (2.0 mL), and 10% LPF (w/w of chestnut oil) as a catalyst, with magnetic stirring for 48 h (300 rpm, 50 ± 2 °C). After that period, the enzyme was filtered, and the filtrate was extracted with dichloromethane (3 × 25 mL). The organic phase was dried over anhydrous Na 2 SO 4 , filtered, and evaporated under reduced pressure. Finally, the expected products were purified via column chromatography with silica gel and a mixture of n-hexane and ethyl acetate (8:2) as the eluent (Barata et al. 2020).

GC-MS analysis
Fatty acids and fatty amides from Bertholletia excelsa oil were characterised with gas chromatography coupled to mass spectrometry (GC-MS), using a Shimadzu/GC 2010 apparatus coupled to a Shimadzu/AOC-5000 autoinjector and an electron beam impact detector (Shimadzu MS2010 Plus) (70 eV), equipped with a DB-5MS fused silica column (Agilent J & WAdvanced 30 m × 0.25 mm × 0.25 mm) (65 kPa). The parameters used were: 1:15 split ratio, helium as the carrier gas, injection volume of 1.0 mL, injector temperature of 250 °C, detector temperature of 270 °C, initial column temperature of 100 °C for 2 min, and rate of heating from 6 °C min −1 to 280 °C for 5 min. The total analysis time was 37 min. Fatty acid amides were identified by comparing their fragmentation spectra with those in the GC-MS library (MS database, NIST 5.0) (Araújo et al. 2018).

Spectroscopic profile in the infrared region
The spectroscopic profile in the infrared region of the samples (Bertholletia excelsa oil and fatty amides) were obtained via impregnation in KBr tablets, and the sample readings were performed on an infrared spectrometer by Fourier transform (FTIR) (Shimadzu IR Prestige -21), with a wavelength of 400-4000 cm −1 and a resolution of 4 cm −1 and 64 scans.

Animals
The zebrafish (Danio rerio) of the wild strain from Aqua New Aquários e Peixe Co. (PE, Brazil) following the methodology of Souza et al. (2016) and Borges et al. (2018) was used. The experiments were carried out following the rules established for the care of animals, and the project was approved by the Ethics Committee on the Use of Animals (CEUA-UNI-FAP) of the Federal University of Amapá (Protocol Number 020/2019).

Determination of the average lethal dose (LD 50 ) and behavioural assessment
To obtain the average lethal dose (LD 50 ), the animals were observed for 48 h after treatment. In this test, adult zebrafish (standard length, 28.1 ± 0.2 mm weighing 0.42 ± 0.04 g) were used. The animals were treated orally through gavage, as described by Carvalho et al. (2017) and Borges et al. (2018). The doses of AGBe administered were 45 mg/kg, 500 mg/kg, and 1000 mg/kg, based on previous studies by de Souza et al. (2020).
Each dose of AGBe was evaluated in groups of four animals. The tests were performed in triplicate, totalling 48 animals (n = 12 per group) distributed in the following groups: group A, vehicle-Thinners (Tween, DMSO, and distilled water) used to solubilize AGBe; group B-AGBe at a dose of 45 mg/kg; group C-AGBe at a dose of 500 mg/kg; group D-AGBe at a dose of 1000 mg/kg; group E-oil of Bertholletia excelsa at a dose of 1000 mg/kg. Behavioural parameters were assessed as described by Souza et al. (2016) and Borges et al. (2018).

Carrageenan-induced abdominal oedema assay
Carrageenan (iota type II, Sigma Co, Lot 65H1096) was administered intraperitoneally (i.p.) in a volume of 20 μL (200 μg) in PBS, according to the methodology described by Borges et al. (2018), 1 h before the oral administration (v.o) of AGBe. Oedema was measured at the maximum peak, which occurred 3 h after the application of carrageenan.
The trial was performed in triplicate, and the doses of AGBe were selected from the toxicity trial and based on previous studies (Collymore et al. 2013;Carvalho et al. 2017;Borges et al. 2018).

Histopathological study
At the end of the toxicological and inflammatory evaluation experiments, the animals were euthanised according to the recommendations of the American Guidelines of the Veterinary Medical Association for Animal Euthanasia (Leary et al. 2013), and the tissues were collected for histopathological analysis.
For histopathological analysis, tissue preparation and microscopic analysis of the organs analysed were based on the techniques described by Souza et al. (2016), Carvalho et al. (2017), and . The histological changes index (IHA) was calculated from the levels of tissue changes observed in the liver, kidneys, and intestines. These alterations were classified into I, II, and III levels, with the IHA value indicating the severity of the organ, 0-10: normal organ; 11-20: mild-to-moderate organ changes; 21-50: moderate to severe organ changes; and > 100, irreversible organ damage, according to Borges et al. (2018) and de Souza et al. (2020).

Statistical analysis
To determine the median lethal dose (LD 50 ), probit analysis was performed using GraphPad Prism software version 6.0. The results of the histopathological study were expressed as mean ± SEM and analysed using ANOVA, followed by the Tukey-Kramer test for comparisons between the treated and control groups. To evaluate anti-inflammatory activity, data were expressed as mean ± standard deviation, and analysed using ANOVA (one-way), followed by the Tukey-Kramer post hoc test. Statistical significance was set at p < 0.05. Data were analysed using the GraphPad Prism software version 8.0.
The genetic optimisation for ligand docking (GOLD) program uses a genetic algorithm for flexible ligand docking experiments within protein binding sites. The GOLD program was used to investigate the modes of interaction between the studied compounds and therapeutic targets (Chandak et al. 2014).
For molecular coupling, hydrogen atoms were added, and the water molecules of the enzymes were removed. Inhibitors complexed with each therapeutic target were then extracted. Before performing the docking simulation, the results were validated by calculating the mean quadratic deviation (RMSD) between the experimental ligand and the conformation of the ligand that produced the best pose after docking.

CG-MS and infrared analysis
In the infrared region of the spectrum, the amides present ( Fig. 2) a band in 3294 cm −1 originating from stretching vibrations of the N-H bond; in 2918 and 2848 cm −1 referring to the asymmetric and symmetrical C-H stretch, respectively; at 1643 cm −1 , an amide group C = O absorption band; in 1558 cm −1 folding band N-H of secondary amides; and at 1468 cm −1 was the observed bands of vibrations of asymmetric angular deformation of the C-H connections of the aliphatic groups.
In the infrared spectrum of the Bertholletia excelsa oil (Fig. 3), it is possible to observe bands in the region of 3007 cm −1 referring to the stretch = −CH; in 2929 and 2854 cm −1 referring to the asymmetric and symmetrical C-H stretch, respectively; in 1747 cm −1 for the C = O stretch and in 1163 cm −1 for the C-O ester stretch. At 1462 cm −1 and 1377 cm −1 , the bands of vibrations of asymmetric angular deformation of the C-H connections of the aliphatic groups can be seen.
The triglyceride characterisation by IR spectroscopy of Bertholletia excelsa oil (Fig. 2), showing a peak in 3007 cm − 1 , corresponds to the stretching of the -C-H The fatty amides from ethanolamine were characterised by MS spectrum with a typical base peak at m/z 116 [for N-(2-hydroxyethyl)oleamide], resulting from McLaffery rearrangement and γ-cleavage, respectively (Fig. 3). In contrast, observations of the ethyl oleate structure mass spectra (Fig. 3) showed the m/z 55 fragmentation ion as the base peak and was less abundant than the ion related to the loss of the ethoxide portion (m/z 264).

Obtaining the average lethal dose (LD 50 )
The animals treated with AGBe and oil of Bertholletia excelsa did not die during or after the experiment, including those treated with the highest dose (1000 mg/kg, v.o). Furthermore, this dose was chosen as AGBe reaches its maximum solubility for oral administration in zebrafish.
The animals showed stressful behaviour in the first hours and soon recovered. Histopathological evaluation showed that both AGBe and Bertholletia excelsa oil did not cause tissue damage that could alter the functioning of the main organs (Fig. 4). In addition, Bertholletia excelsa oil was not toxic to the liver (Fig. 5).

Carrageenan-induced abdominal oedema assay
The application of carrageenan to the animals' peritoneum produced visible oedema, with a maximum peak observed in the third hour after application (Fig. 6A), which was inhibited by around 48% in the spilantol group. Treatment with AGBe at doses of 100, 500, and 750 mg/kg produced a dose-dependent effect, with the highest dose producing 95% inhibition (Fig. 6B). The effects of the treatments are visible macroscopically in Fig. 7.
The IHA values of the groups treated with AGBe demonstrated that the liver (5.66 ± 0.233), kidney (3.41 ± 0.258), and intestine remained functionally normal. The group treated with spilantol at a dose of 35 mg/kg produced changes in the intestine with IHA of 12.66 ± 0.05, which was significantly different from the control group (Fig. 8), with marked intestinal histopathological changes, mainly the occurrence of goblet cell hyperplasia (Fig. 10). In the groups treated with AGBe, the observed alterations were not decisive for the loss of tissue function (Figs. 9, 10, 11).
The 18:0-ethanolamine fatty amide showed 7 interactions, 6 of which were hydrophobic and 1 of hydrogen with HIS90, LEU352, PHE518, MET522, and ILE523-amino acids with a score value for the best position of 74.41.
Considering the amino acids from the active site, the interactions of the spilantol molecules, 18:2, ω-6-ethanolamine, 16:0-ethanolamine, and 18:1, ω-9-ethanolamine, with the TYR385 residue were observed. Moreover, for the SER530 amino acid residue, only the 18:1 ω-7-ethanolamine molecule interacted. This last molecule is noteworthy, because it has the most significant number of interactions and, even though it does not interact with the TYR385 residue, it did interact with a LEU384 nearby residue. The 18: 0-ethanolamine molecule showed no interaction with the amino acids of the active site.
Considering the amino acids from the active site, all molecules interacted with the PHE5 amino acid residue. For the ILE9 residue, only spilantol, 16:0-ethanolamine, 18:2, ω-6-ethanolamine, 18:1, and ω-9-ethanolamine showed intermolecular interactions. The 18:1 ω-9-ethanolamine molecule stands out for the PLA 2 target with the highest score value and number of interactions, and interacts with the two amino acid residues of the active site.

Discussion
In recent years, zebrafish have become the most used animal in laboratories, increasing their implementation in experiments and scientific research. Most of the research results are gathered in the ZFIN community-Zebrafish Information Network (Spence et al. 2008).
Zebrafish are thus replacing the use of rodents owing to several favourable characteristics. According to Souza et al. (2016), one of them is its small size, which makes the dose administered, whether in microliters (µL) or less, of the active ingredient induces a pharmacological effect. It is highly favourable for the testing of natural products, because the material used to obtain the product is not always abundant, and the extraction efficiency can be very low. In addition, 70% of genes orthologous to Homo sapiens are highly similar to humans in terms of function of the main physiological processes, with key organ systems, such as the digestive, nervous, and cardiovascular systems (Hsu et al. 2007). It largely favours the equivalence of the response to pharmacological agents between the two species (MacRae and Peterson 2015).
Innovative studies have introduced a new route of administration to assess the toxicity of possible drugs. Oral treatment (gavage) is an efficient method for assessing the toxic potential of natural or chemical substances. The histopathology of the liver, kidney, and intestinal tissues has also been evaluated (Collimore et al. 2013;Borges et al. 2018;Sampaio et al. 2018).
Animals, when in contact with any toxic substance, whether of synthetic or natural origin, may present characteristics that indicate a possible toxic effect in the short or long term. According to Ribeiro (2013), these substances can trigger changes in different systems and behaviour, and can even cause death in animals (Mathur et al. 2011).
According to Huang et al. (2014), Borges et al. (2018), and de Souza et al. (2020), zebrafish, when in contact with foreign substances, adopt patterns of stress behaviour, as observed in this study after the oral administration of AGBe; however, the studied substances did not cause damage to the tissue level of the organs analysed in the histopathological study.
In the toxicity test, it was observed that Bertholletia excelsa oil and AGBe did not show toxicity with oral treatment (1000 mg/kg), and it was not possible to determine the LD 50 . It is noteworthy that we decided to use this dose above the effective dose to ensure the safety of AGBe administration. A similar result was obtained by Barata et al. (2020), who reported reduced cell toxicity for fatty amides. de Souza et al. (2020) reported that even substances of natural origin that do not cause behavioural changes or death in the zebrafish could cause internal damage in this animal, altering the normal functioning of some organs. According to the parameters presented in studies carried out by Souza  , the rate of histopathological changes observed in this study for the kidneys and intestines of animals treated with AGBe and Bertholletia excelsa oil were normal, as they did not present changes that compromised the homeostatic pattern of the organs.
In this study, Bertholletia excelsa oil was not toxic to the liver, and the IHA was 0. This result reinforces the findings of Pawel et al. (2013) and Barata et al. (2020), who did not demonstrate toxicity to the liver and kidneys in rats, and also showed low cellular toxicity for fatty amides. Carnovali et al. (2016) evaluated the action of fatty acid amides in zebrafish and demonstrated that they prevent the alteration of bone markers in a prednisolone-induced osteoporosis model in adult zebrafish scales, whereas their esterified forms did not. These data suggest that long-chain fatty acid amides are involved in regulating bone metabolism.
In this study, carrageenan was used as an inflammatory agent in a zebrafish model. Huang et al. (2014) validated the use of carrageenan as an inflammatory inducer in the zebrafish peritoneum and observed that i.p. injection of carrageenan produced typical symptoms of inflammation, such as swelling, and upregulated MPO, a leukocyte marker, as well as the pro-inflammatory proteins TNF-α and iNOS. They also demonstrated that local injection of carrageenan into soft tissues induces acute inflammation, and that known compounds with anti-inflammatory properties can modulate the inflammatory responses of carrageenan-injected adult zebrafish.
Thus, for the evaluation of AGBe in the inflammatory process triggered by carrageenan, the protocols of Huang et al. (2014), Carvalho et al. (2017), and Borges et al. (2018) regarding intraperitoneal carrageenan were used to induce the formation of abdominal oedema in zebrafish.
The participation of cyclooxygenase products (prostaglandins) in carrageenan oedema, especially in the second phase, has already been described in several studies (Zaa et al. 2012;Motta et al. 2013;Huang et al. 2014;Carvalho et al. 2017;Borges et al. 2018;Barata et al. 2020). In addition, non-steroidal anti-inflammatory drugs, such as indomethacin, are inhibitors of prostaglandin synthesis via COX-1 inhibition and IL-6 production (Motta et al. 2013).
The administration of carrageenan intraperitoneally produced the formation of abdominal oedema, which was more visible in animals treated with thinner/carrageenan (Fig. 6B), and treatment with different doses of AGBe (100, 500, and 750 mg/kg) orally produced an inhibitory effect on carrageenan oedema in a dose-dependent manner (Fig. 8B). These results align with those described by Barata et al. (2020) for amides obtained from triglycerides of Bertholletia excelsa oil, which demonstrated antioedematogenic activity on rat paw carrageenan oedema.
The fact that fatty acid amides are described as inhibitors of cyclooxygenase and lipoxygenase (Fiorucci et al. (2001) and Barata et al. (2020)), in this study we consider to support the effect of AGBe.
In this study, molecular docking was performed for AGBe and the standard anti-inflammatory drugs. This computational method is currently widely used to obtain new drugs (Du et al. 2016). It describes the mode of interaction of molecules at the enzyme or receptor site through specific fundamental interactions and predicts the binding affinity between protein-ligand complexes. Spilantol was used as a standard in the in silico study to compare the results as it has a chemical structure similar to the studied molecules and presents a report in the literature on anti-inflammatory activity (Wu et al. 2008).
The RMSD value indicates the accuracy of the docking poses calculated by the GOLD fitting algorithm compared to the experimentally determined poses for a compound linked to a biological target. Therefore, the calculation of docking with an RMSD of less than 2 Å for a proper conformation is considered successful. Therefore, it has justified validity (Cole et al. 2005).
Prostaglandins are derived from arachidonic acid (AA) in a reaction catalysed by COX, which can exist as COX-1 and COX-2. AAs are released from the cell membrane upon neopathological stimuli. Inhibitors of this enzyme interfere with this reaction, and the disease process begins. Recently, the involvement of COX-1 in cancer and inflammation has been firmly established (Vitale et al. 2016;Hage-Melim et al. 2019). Fig. 15 Docking of compounds Spilantol, 16:0-ethanolamine, 18:2, ω-6-ethanolamine, 18:1, ω-9-ethanolamine, 18:1, ω-7ethanolamine e 18:0-ethanolamine performing interaction with PLA 2 Chunhieng et al. (2008) and Barata et al. (2020) confirmed that the polyunsaturated fatty acids present in the oil of Bertholletia excelsa have different fatty amide precursors, which have anti-inflammatory properties and probably act in the COX pathway, as was observed in this study. AGBe identified as 18: 1, ω-7-ethanolamine and 18: 1, ω-9-ethanolamine, present in vaccenic and oleic fatty acids, showed more significant interaction for COX-2 and PLA 2 , which stood out for presenting a score and number of interactions greater than the spilantol pattern, interacting with the amino acids present in the active site or, at least, close to it in all the studied targets ( Fig. 15, 16, and 17).
The AGBe in the docking between the therapeutic targets (Figs. 12 and 13) presented a higher score for the COX-1 therapeutic target, with interactions in important amino acids of this enzyme, with the 18: 1, ω-7-ethanolamine amide, presenting the highest score value of 78.40.
Spilantol was used as a standard for comparison, because it has a chemical structure similar to AGBe, and because it has anti-inflammatory activity (Wu et al. 2008) and, with the therapeutic target COX-2 (Fig. 14), spilantol and 18:2, ω-6-ethanolamine showed interactions, with amino acids with score values of 63.00 and 76.71, respectively. With the therapeutic target PLA 2 (Fig. 15), spilantol had a score value of 62.45 and the 18: 1 molecule, ω-9-ethanolamine, had the highest score value and several interactions and interacted with the two amino acid residues of the active site.
Prostaglandins are derived from arachidonic acid (AA) in a reaction catalysed by COX, which can exist as COX-1 and COX-2. After neopathological stimuli, AA is released from the cell membrane. Inhibitors of this enzyme interfere with this reaction, and the involvement of COX-1 in cancer and several inflammatory processes is already known (Vitale et al. 2016;Melim et al. 2019).
The RMSD value indicates the accuracy of the docking poses calculated by the GOLD fitting algorithm compared to the experimentally determined poses for a compound linked to a biological target. Thus, the calculation of docking with an RMSD of less than 2 Å for a conformation of fit is considered successful. Therefore, it has justified validity (Cole et al. 2005). Therefore, all AGBe studied had a score value and number of interactions greater than the standard molecule (spilantol), indicating anti-inflammatory activity related to COX-2 and PLA 2 inhibition. Carvalho et al. (2017) demonstrated that the administration of an inflammatory agent, such as carrageenan, in the abdominal region of Danio rerio can cause reactions in vital organs such as the gills, liver, intestine, and kidneys. Borges et al. (2018) stated that the technique of intraperitoneal injection in zebrafish is invasive, which can easily cause damage to the organs contained in the abdominal cavity responsible for the metabolism and excretion of various substances.
The histopathological study observed that the group treated with spilantol ( Fig. 8) had an IHA of 12.66 for the intestine, considering mild-to-moderate changes. de Souza et al. (2020) reported that spilantol, depending on the dose, can influence the production of histopathological damage in the intestine, liver, and kidneys in zebrafish and reported that spilantol caused irreversible damage to the intestines of animals. In this study, the group treated with AGBe at the highest dose (750 mg/kg) and received carrageenan as an oedematogenic agent, did not present histopathological alterations in organs evaluated that could compromise the physiological functions (Figs. 9, 10, and 11), and presented 95% inhibition of the inflammatory process triggered by carrageenan in the zebrafish peritoneum (Fig. 7B). This fact may be related to the modulation of pathophysiological mechanisms triggered by carrageenan, highlighting the participation of prostaglandins in the maximum peak of oedema (Borges et al. 2018). The anti-inflammatory action hypothesis was confirmed in the in silico study, demonstrating the involvement of AGBe in the process of inhibiting the enzymes COX-2 and PLA 2 .

Conclusion
The method used to obtain AGBe from Bertholletia excelsa oil was effective and, considering the results obtained in the carrageenan oedema test in zebrafish, it can be suggested that AGBe has anti-inflammatory activity, including triggering a dose-response effect. The hypothesis of antiinflammatory action was confirmed in the in silico study, demonstrating the involvement of AGBe in inhibiting the enzymes COX-2 and PLA 2 , with emphasis on the molecules 18: 1 ω-7-ethanolamine and 18: 1, ω-9-ethanolamine. In the histopathological study, AGBe did not cause significant changes to the main metabolising organs (liver, kidneys, and intestines), whereas spilantol produced mild-to-moderate changes in the intestinal tissue. Therefore, based on all the results obtained and the fact that until the dose of 1000 mg/ kg, orally, in zebrafish, it was not possible to determine the LD 50 , it can be said that AGBe is effective and safe for antiinflammatory activity.