GYA Grasas y Aceites Grasas y Aceites 1988-4214 0017-3495 Consejo Superior de Investigaciones Científicas gya.0102211 10.3989/gya.0102211 Research Chemical-functional composition of Terminalia catappa oils from different varieties Composición química-funcional de aceites de Terminalia catappa de diferentes variedades https://orcid.org/0000-0001-5423-1945 Santos O.V. Universidade Federal do Pará - UFPA. Rua Augusto Correa, 1, Bairro: Guamá, 66075-110, Belém, Pará, Brazil. https://orcid.org/0000-0002-4229-7805 Soares S.D. Universidade Federal do Pará - UFPA. Rua Augusto Correa, 1, Bairro: Guamá, 66075-110, Belém, Pará, Brazil. https://orcid.org/0000-0002-2807-711x Dias P.C.S. Universidade Federal do Pará - UFPA. Rua Augusto Correa, 1, Bairro: Guamá, 66075-110, Belém, Pará, Brazil. https://orcid.org/0000-0002-5235-6880 Duarte S.P.A. Universidade Federal do Pará - UFPA. Rua Augusto Correa, 1, Bairro: Guamá, 66075-110, Belém, Pará, Brazil. https://orcid.org/0000-0001-6673-4419 Santos M.P.L. Universidade Federal do Pará - UFPA. Rua Augusto Correa, 1, Bairro: Guamá, 66075-110, Belém, Pará, Brazil. https://orcid.org/0000-0002-2817-3312 Nascimento F.C.A. Universidade Federal do Pará - UFPA. Rua Augusto Correa, 1, Bairro: Guamá, 66075-110, Belém, Pará, Brazil. https://orcid.org/0000-0003-3695-7499 Teixeira-Costa B.E. betcosta@ufam.edu.br Universidade Federal do Amazonas - UFAM. Avenida General Rodrigo Octavio, 1200, Coroado I, 69077-000, Manaus, Amazonas, Brazil. Programa de Pós-Graduação em Ciência de Alimentos, Instituto de Química, Universidade Federal do Rio de Janeiro - UFRJ, Avenida Athos da Silveira Ramos, 149, 21941-909, Rio de Janeiro, RJ, Brazil. 22 05 2022 06 2022 73 2 e454 03 01 2021 17 05 2021 13 06 2022 ©2022 CSIC 2022 This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International (CC BY 4.0) License. Summary

This study aimed to extract and physical-chemically characterize Terminalia catappa L. kernel oil from purple (CR) and yellow (CA) varieties. Physical-chemical parameters, composition of fatty acids, nutritional quality indices, bioactive compounds and antioxidant capacity of both oil varieties were evaluated according to the literature. Both oils presented low levels of acidity and peroxides, besides the predominance of unsaturated fatty acids, ~63% of oleic and ~26% of linoleic acids, which influenced its nutritional indices. The CR oil variety exhibited a higher content in anthocyanin (18.3 ± 1.5 mg·100 g-1), ascorbic acid (68.4 ± 2.02 mg·100 g-1) and total polyphenol contents (152.3 ± 2.4 mg GAE·g-1), and a good antioxidant activity (38.6 ± 2.2 μg TE·g-1) determined by TEAC assay, when compared to the CA oil (p < 0.05). Therefore, the results confirm the importance of T. catappa as a lipid source for human consumption to be used in the development of food products.

Resumen

El objetivo de este estudio fue extraer y caracterizar físico-químicamente aceite de semilla de Terminalia catappa de las variedades violeta (CR) y amarilla (CA). Se evaluaron parámetros fisicoquímicos, composición de ácidos grasos, índices de calidad nutricional, compuestos bioactivos y capacidad antioxidante de ambas variedades de aceite de acuerdo con a la literatura. Como resultado, ambos aceites presentaron bajos niveles de acidez y peróxidos, y predominio de ácidos grasos insaturados, ~63% de ácido oleico y ~26% de ácido linoleico, lo cual influyó en su perfil nutricional. La variedad de aceite CR presentó un mayor contenido de antocianina (18,3 ± 1,5 mg·100 g-1), ácido ascórbico (68,4 ± 2,02 mg·100 g-1) y contenido total de polifenoles (152,3 ± 2,4 mg GAE·g-1), y una alta actividad antioxidante (38,6 ± 2,2 μg TE·g-1) determinado por ensayo TEAC, en comparación con el aceite CA (p<0.05). En conclusión, los resultados presentados refuerzan la importancia de T. catappa como fuente de lípidos para la ingesta humana y para su uso en el desarrollo de productos alimenticios.

 Antioxidant activity Bioactive substances Linoleic acid Oleic acid Tropical almond Vegetable oil Aceite vegetal Ácido linoleico Ácido oleico Actividad antioxidante Almendra tropical Sustancias bioactivas Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) 001 The authors acknowledge Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) Finance Code 001. Introduction

Among the diversity of fruit species in Brazil there are underexploited species that can be used for human nutrition, as well as for the extraction and isolation of functional/bioactive compounds, which can play an important role in maintaining human health. In this context, a native species from tropical and subtropical regions, Terminalia catappa L., stands out as an innovative source of fruits and their derivatives. It belongs to Combretaceae family, and produces glabrous, rounded and flattened drupaceous fruits. These fruits are commonly named Sea-almond, Tropical-almond, Indian-almond or Malabar-almond (Abdulkadir, 2015).

The Tropical almond fruits initially exhibit a green color, which during the maturation process becomes red-purple and, also may turn yellow (Salawu et al., 2018). The fruits measure 5-7 cm long and 3-6 cm in width, have an exocarp (bark) adhered to an edible fibrous pulp (mesocarp), and a single rigid seed. The fruit pulp is a good source of carbohydrates, up to 76%, and low in lipids, less than 3% (Ladele et al., 2016). The seed has an oily endocarp (kernel), containing up to 52% lipids, 38% proteins and minerals, which is covered by a thin peel (Agu et al., 2019; Abdulkadir, 2015; Souza et al., 2016). Both pulp and seeds are edible (Ladele et al., 2016). The lipids extracted from the kernel almond have great potential for application as an edible vegetable oil due to their elevated lipid yield, up to 60%, which has higher value when compared to main commercial oilseeds, such as soybeans, palm, and peanuts (Ladele et al., 2016; Jokić et al., 2015).

Terminalia catappa can be considered a fruit tree with high economic potential because its fruiting starts at around 3 to 5 years of age, with two harvests per year, producing up to 30 kg fruits per year, reaching an estimated world production of more than 700,000 tons in 2004 (Agu et al., 2019; Singh and Choudhary, 2012). However, its productivity needs further assessment. In addition, this species is commonly cultivated in several countries as an ornamental tree, a fruit tree and as a vegetable oil source for different applications (Agu et al., 2019; Janporn et al., 2015; Menkiti et al., 2015). The Tropical almond cultivation generally needs low maintenance, since it has a simple propagation from seeds and can grow quickly in different soils and environments (Ladele et al., 2016).

Investigations have been carried out to characterize the Tropical almond fruit and to determine and quantify its bioactive compounds as a source of natural antioxidants (Abdulkadir, 2015; Huang et al., 2018). These bioactive compounds have been studied for complementary functions and actions of insulin in the treatment of diabetes, to act regulating dietary constituents in human daily intake and as potential anti-inflammatory agents (Ben et al., 2019; Huang et al., 2018). A recent study has focused on the nutritional and functional properties of the pulp and kernel oils of Terminalia catappa L. obtained by supercritical fluids (Santos et al., 2021). In another work from Agu et al. (2019), the T. catappa kernel oil was chemically modified and characterized as a potential replacement for mineral transformer fluid. In a similar application, the oil from T. catappa was used by Silva et al. (2020b) to synthetize biodiesel (via methyl route).

It is worth mentioning that in the works in the literature, a few of them have identified which variety of the Tropical almond, purple or yellow, was used in their research. Thus, investigations comparing different varieties of Tropical almond fruit can increase knowledge based on its different chemical compositions and bioactive/nutritional constituents, and guide new applications for the food and chemical industries.

The aim of this research was to evaluate the functional chemical composition of Terminalia catappa L. kernel oil, and to compare its purple (CR) and yellow (CA) varieties. Fatty acids and triacylglycerol profile, nutritional quality parameters, bioactive compounds, such as anthocyanins, ascorbic acid and polyphenol contents, and antioxidant activity were investigated.

Materials and methods Raw material and oil extraction

Fruit seeds of Terminalia catappa L. from purple (CR) and yellow (CA) varieties were harvested on the campus of Federal University of Pará (UFPA) in 2020. The fruits were collected in the following geographical coordination: latitude 01º 27 ‘21”S, longitude 48º 30’ 16”W and altitude 10 m. To ensure the proper taxonomic identification of this plant, some parts of it, such as leaves and fertile material, were collected and deposited in the herbarium Professor Normélia Vasconcelos/UFPA under the code MG nº 3791.

Sampling and oil extraction

The fruits of Terminalia catappa L. from purple and yellow varieties were washed to remove any physical dirt, gently peeled off, and then the seeds were manually cracked, and their kernels were removed. The kernels were dried at 60 °C for 24 h in an air-circulation oven (model 81-150, New Lab Equipamentos, Piracicaba, SP, Brazil), and milled in a Willey miller (TE-650 model, Tecnal, SP, Brazil). Subsequently, a solid-liquid extraction was carried out in a Soxhlet apparatus, using hexane as solvent according to the methodology of Silva et al. (2020b). All analyses were performed in triplicate. The oil yield (%) was calculated according to the Eq. 1.

O i l   y i e l d   % = W o i l W T g × 100

where Woil is the extracted mass of oil (g) and the Wg is the total mass of seeds (g).

 Physical-chemical analysis of T. catappa kernel oils Quality parameters

The physical-chemical quality parameters of the Terminalia catappa L. kernel oils from purple (CR) and yellow (CA) varieties were determined according to the official methods from the American Oil Chemist’s Society (AOCS), as follows: acidity, peroxide and saponification values were determined according to AOCS methods Cd 3d-63, Cd 8-53 and Cd 3-25, respectively (AOCS, 2004). The true density (ρ, g m-3) was measured using a DA-130 digital density meter (Kem Kyoto Electronics, Japan) at room temperature (~25 °C) and the refractive index was investigated according to the Cc 7-25 method (AOCS, 2004).

Fatty acids profile

The fatty acid (FAs) profile of Terminalia catappa L. oils CR and CA was determined as fatty acid methyl esters (FAMEs) according to the established procedure ISO 5509:2000 reported by the International Organization for Standardization (ISO, 2000). After phase separation, the supernatant was collected for subsequent gas chromatographic analysis with flame ionization detector (GC-FID) (Thermo Scientific Trace GC Ultra) using a wall-coated open-tubular column (WCOT). The analysis was performed in a gas chromatograph (Varian 430 model, Agilent Technologies, CA, USA) equipped with a microcomputer with the software Galaxie Chromatography under the following parameters: fused silica SP®-2560 capillary column (Merck, USP-G5, SUPELCO, USA) of 100 m in length and 0.25 mm internal diameter, containing 0.2 μm of polyethylene glycol. The operation conditions were: 50:1 split injection ratio, column temperature at 140 °C for 5 min programmed with an increasing rate of 4 °C·min-1 up to 240 °C, helium as carrier gas in 37 psi isobaric pressure, 20 cm·sec-1 linear velocity, make up gas: 29 mL·min-1 helium flow, 250 °C injector temperature, autosampler model Varian CP8410, detector temperature 250 °C. The peaks were identified by comparing peak retention time to the known FAMEs standard (37-Component FAME Mix - methyl esters of fatty acids ranging from C4 to C24 CRM47885, Supelco). The quantitative composition was carried out by area normalization, and expressed in mass percentage as established by the official method Ce 1-62 (AOCS, 2004). The samples were analyzed in triplicate.

 Nutritional quality indices

The nutritional quality indices in the Terminalia catappa L oils from purple and yellow varieties were established based on their respective FAs profiles, which were classified according to the presence and number of double or triple bonds: saturated fatty acids (SFA), unsaturated fatty acids (UFA), monounsaturated fatty acids (MUFA) and polyunsaturated fatty acids (PUFA). The following indices were used to investigate its nutritional quality: atherogenicity (AI) and thrombogenicity indices (TI) were determined according to Ulbricht, Southgate (1991), and calculated according to Eq. 2 and Eq. 3, respectively. The hypocholesterolemic/hypercholesterolemic ratio (h/H) was determined as defined by Santos-Silva et al. (2002) and calculated using the Eq. 4. The calculated oxidative stability value (COX) was defined according to Silva et al. (2020a) as expressed in the Eq. 5.

A I =   ( C 12 : 0 +   4   × C 14 : 0 + C 16 : 0 ) ( P U F A + M U F A ) T I = ( C 12 : 0 + C 16 : 0 + C 18 : 0 ) 0.5   × M U F A + 0.5   × n 6 - P U F A + 3   × n 3 - P U F A + n 3 - P U F A n 6 - P U F A h / H = C 18 : 1 + P U F A ( C 14 : 0 + C 16 : 0 ) C O X = C 18 : 1 + 10.3 × C 18 : 2 + ( 21.6 × C 18 : 3 ) 100 Triacylglycerol composition

The triacylglycerol composition in the Terminalia catappa L. oils from purple and yellow varieties was estimated based on the 1,3-random-2-random distribution hypothesis using the software PrOleos®, which predicts the molar percentage of triacylglycerols present in oil based on its fatty acid composition (Antoniosi Filho et al., 1995). This software is available online at the website “https://lames.quimica.ufg.br/p/4035-apoio-didatico”.

 Analysis of bioactive compounds and antioxidant capacity

The bioactive compounds and antioxidant activity in Terminalia catappa L. kernel oils from purple and yellow varieties were investigated. The bioactive compounds were analyzed as anthocyanin, ascorbic acid and total polyphenol contents, and the antioxidant activity was determined based in the Trolox Equivalent Antioxidant Capacity (TEAC) assay. Prior the analysis, the oil samples were solubilized in isopropyl alcohol at 320 mg·mL-1 concentration.

Anthocyanin content. The content of anthocyanins was determined as reported by Silva et al. (2014). About 1 g of each sample was mixed with 10 mL of a 1.5N HCl in 85% ethanol solution. The samples were homogenized and left to rest overnight under refrigeration and dark covered. Then, the absorbance of the samples was measured at 535 nm wavelength using an UV-Vis spectrophotometer (model UV-1800, Shimadzu, Tokyo, Japan). The analysis was performed in triplicate. The anthocyanins’ content was calculated using the Eq. 6 and results were expressed as mg·100g-1.

A n t h o c y a n i n   c o n t e n t = A b c   × d i l u t i o n   f a c t o r s × 1000 W s a m p l e × ε 1 c m , 535 1 %

where Abs is the measured absorbance of the sample at 535 nm, Wsample is the weight of the sample and ε 1 c m , 535 1 % is the absorption coefficient for anthocyanins, which is equal to 982 g·100 mL-1cm-1.

Ascorbic acid content. The ascorbic acid content was determined by the reduction of the 2,6-dichlorophenol-indophenol compound, according to the adapted methodology of Cunha-Santos et al. (2019). About 10 mL of sample were mixed with 2 mL of a 0.03g·mL-1 metaphosphoric acid diluted in an acetic acid aqueous solution and titrated with 0.2% 2,6-dichlorophenol-indophenol solution with sodium bicarbonate at 0.21 mg·mL-1 concentration, until the appearance of a pink color was persistent for more than 5 s. The 2,6-dichlorophenol-indophenol solution was standardized with an ascorbic acid solution prior to analysis. The sample was analyzed in triplicate and the results were expressed as mg of ascorbic acid per 100 g of sample (mg·100g-1).

Total polyphenol content. The total polyphenol content of these fractions was analyzed following the Folin-Ciocalteu assay as reported by Aliakbarian et al. (2011). Initially, 0.2 mL sample, 4.8 mL deionized water, and 0.5 mL Folin-Ciocalteu reagent (Sigma-Aldrich) were transferred to a 10 mL volumetric flask, and vigorously mixed. Then, 1 mL of a 20% sodium carbonate solution was added, followed by deionized water until reaching a final volume of 10 mL. The solutions were mixed and left to rest at room temperature in the dark for 1 h. An aliquot (~2 mL) of sample was used for the determination of total polyphenols using a UV-Vis spectrophotometer (model UV-1800, Shimadzu, Tokyo, Japan) at a wavelength of 725 nm. Distilled water was considered as blank. The sample was analyzed in triplicate, and the results were calculated based on a standard curve of gallic acid (Sigma-Aldrich) and expressed as mg GAE·g-1.

Antioxidant capacity. The determination of the antioxidant capacity from the samples was performed according to the Trolox equivalent antioxidant capacity (TEAC) assay using the ABTS (2,2’-azinobis 3-ethylbenzthiazoline-6-sulfonic acid, from Sigma-Aldrich) reagent as described by Chen et al. (2011). The absorbance was measured at 734 nm wavelength using UV-Vis spectrophotometer (model UV-1800, Shimadzu, Tokyo, Japan). The assay was performed in triplicate against a calibration curve of Trolox (μg Trolox equivalent·L-1) and calculated using the following linear equation:

T E A C = ( 0.6239 - A b s o r b a n c e 734 n m ) / 0.3364     ( R 2 = 0.997 ) Statistical analyses

The results were statistically analyzed using the Statistica software version 7.0 (Statistica, 2000), by analysis of variance (ANOVA) and Tukey’s test at the significance level of 5% (p < 0.05).

 Results and discussion Oil extraction

The solid-liquid extraction of the Terminalia catappa L. kernel oils from purple (CR) and yellow (CA) varieties showed a lipid yield of 57 and 54%, respectively. These results are higher than the value of 52% found by Silva et al. (2020b) and lower than the value of 61.7% obtained by Ladele et al. (2016) for kernel oil yield from Terminalia catappa seeds. It is worth mentioning that in both studies, hexane was used as solvent for solid-liquid extraction, and the seeds were harvested in Brazil and Benin, respectively.

Physical-chemical analysis of the T. catappa kernel oils Quality parameters

The results from the quality parameters of the Terminalia catappa L. kernel oils from purple (CR) and yellow (CA) varieties are shown in Table 1. The quality parameters in vegetable oils, acidity, and peroxide values, are ruled by the Codex Alimentarius (2001). This institution recommends the maximum values for acidity and peroxide, in crude vegetable oils as 4 mg KOH·g-1 and 15 meq·Kg-1, respectively.

Quality parameters of <italic>Terminalia catappa</italic> kernel oils from purple (CR) and yellow (CA) varieties.
Quality parameters Oil samples
CR CA
Acidity value (mg KOH·g-1) 1.25 ± 1.05a 1.55 ± 0.47a
Peroxide value (meq O2·Kg-1) 2.05 ± 1.17a 3.43 ± 0.72a
Saponification value (mg KOH·g-1) 185.5 ± 0.18a 180.7 ± 1.05a
Refractive index 1.50 ± 0.01a 1.45 ± 0.00a
Density (g·m-3) 0.91 ± 0.00a 0.90 ± 0.00a

Data represent the mean ± standard deviation of triplicate analyses (n = 3). Different superscript lowercase letters in the same line represent significant differences (p < 0.05) at 95% confidence interval according to Tukey’s test.

The acidity values for the CR and CA oils are in accordance with the Codex Alimentarius (2001) standard values for crude vegetable oils. When comparing this result to the acidity value for T. catappa kernel oil from Benin and Congo, ~2.24 and ~2.42 mg KOH·g-1 respectively, from the work of Ladele et al. (2016), it was observed that the CR and CA oils presented a lower value. In the work of Janporn et al. (2015), the acidity value for the T. catappa oil from Thailand was around 2.4 mg KOH·g-1, a higher value when compared to the CR and CA oils. The determination of acidity in vegetable oils is an important indicator of the presence of free fatty acids, which can be associated with lipid hydrolytic degradation and quality loss (Ghafoor et al., 2019).

The peroxide values determined for CR and CA oils were below the maximum value recommended by the Codex Alimentarius (2001), demonstrating its good quality. It was observed that the CA oil presented a higher peroxide value than CR, although with no statistically significant difference (p > 0.5). When investigating the peroxide value of T. catappa oils from Benin, Nigeria and Congo extracted using organic solvents, Ladele et al. (2016), found similar values of 3.7, 2.8 and 0.5 meq O2·Kg-1, respectively. In another work, the crude oil of T. catappa from Thailand presented a lower value for peroxides at 0.65 meq O2·Kg-1 (Janporn et al., 2015). The peroxide value is a crucial factor in the quality evaluation of edible oils as it can be correlated to the presence of secondary lipid oxidation products and may cause rancidity. Besides, it is well established that high temperatures during processing, storing, as well as long-time exposures to light, humidity and atmospheric oxygen are key factors to lipid oxidation, which is reflected in high levels of acidity and peroxides.

The saponification value for T. catappa CR and CA oils was higher than the amount of ~175 mg KOH·g-1 obtained by Ladele et al. (2016), and ~179 mg KOH·g-1 as determined in the work of Janporn et al. (2015). When comparing both varieties of T. catappa oils, purple and yellow, the first one was found to present a higher saponification value, although with no significant difference (p > 0.05). In the Codex Alimentarius (2001) there is no indication of maximum value for saponification in crude vegetable oils, but there is a recommended value for virgin olive oil of 184 - 196 mg KOH·g-1. When considering these limits, the CR and CA oils presented lower values, which is a good indication of quality. The saponification value is commonly used to estimate the average length of FA chains, which may indicate a high percentage of short-chain ester bonds and a higher saponification value.

The physical properties of vegetable oils, such as density, refractive index, viscosity, and other rheological parameters are factors to be considered when considering the processing design of equipment, as well as its proper function, e.g., pumping, settling and filtration (Freitas et al., 2018). Nevertheless, the density and the refractive index of CR and CA are in accordance with the literature (Ghafoor et al., 2019; Freitas et al., 2018). For both results, the CR and CA oils presented no significant difference (p > 0.5) between each other. Furthermore, the quality parameters of CR and CA oils are in accordance with the literature and international standards (Codex Alimentarius, 2001).

 Fatty acid profile and nutritional quality indices

The composition on FAs of Terminalia catappa kernel oils from purple (CR) and yellow (CA) varieties is shown in Table 2, and for comparison purposes the FA profiles of authentic vegetable oils from maize, soyabean and palm kernel determined by the Codex Alimentarius (2001) were listed. The FA profiles of CA and CR oils exhibited the predominance of unsaturated fatty acids (UFAs), up to 62.9%, mainly represented by oleic acid. The major proportions of FAs in both samples were oleic acid (up to 39%), follow by palmitic acid (~33%), then linoleic acid (~26%). When comparing the CR and CA oils, the percentages of oleic and linoleic acids presented significant differences (p < 0.05), a behavior that was not observed for the other FAs. Furthermore, the FA profiles of CR and CA are in accordance with the literature. In the work of Janporn et al. (2015), the oil from T. catappa, extracted by solvent using the Soxhlet apparatus, presented a remarkably similar FA profile. For these authors, the major proportions of FAs were oleic acid (~31.7%), followed by palmitic acid (~31.4%), then linoleic acid (~23%). The composition of FAs in the T. catappa kernel oil from Benin, investigated by Ladele et al. (2016), also displayed a similar profile, in which the palmitic (~40%), linoleic (~26.6%) and oleic acids (~26.2%) stood out.

Comparative profiles of fatty acids in <italic>T. catappa</italic> kernel oils from purple (CR) and yellow (CA) varieties and other oilseeds.
% Total fatty acids Terminalia catappa oils Edible vegetable oils*
CR CA Maize Soyabean Palm kernel
Myristic acid (C14:0) 0.10 ± 0.00 --- < 0.3 < 0.2 0.5−1.5
Palmitic acid (C16:0) 31.08 ± 0.00a 33.4 ± 0.9a 8.6−16.5 8−13.5 38−43.5
Palmitoleic acid (C16:1) 0.38 ± 0.00a 0.29 ± 0.00a < 0.5 < 0.2 < 0.6
Stearic acid (C18:0) 5.72 ± 0.01a 5.62 ± 0.01a <3.3 2−5.4 3.5−5
Oleic acid (C18:1 cis ω-9) 39.08 ± 0.02a 33.9 ± 2.3b 20−42.2 17−30 39.8−46
Linoleic acid (C18:2 cis ω-6) 22.80 ± 0.03a 26.0 ± 2.6b 34−65.6 48−59 10−13.5
α-linolenic acid (C18:3 ω-3) 0.06 ± 0.01a 0.04 ± 0.05a <2 4.5−11 <0.6
Arachidic acid (C20:4 ω-6) 0.62 ± 0.00a 0.55 ± 0.01a ---- ---- ----
Behenic acid (C22:0) 0.19 ± 0.01a 0.16 ± 0.06a <0.5 ---- ----
Σ SFAs 37.10 39.20 ---- ---- ----
Σ UFAs 62.90 60.80 ---- ---- ----
Σ MUFAs 39.46 34.20 ---- ---- ----
Σ PUFAs 23.50 26.70 ---- ---- ----
Σ ω-6 23.42 26.65 ---- ---- ----
Σ ω-3 0.06 0.04 ---- ---- ----
Total 100.00 100.00 ---- ---- ----

*Values determined from authentic samples by Codex Alimentarius (2001). ---- = Non-defined. Data represent the mean ± standard deviation of triplicate analyses (n = 3). Different superscript lowercase letters in the same line represent significant differences (p < 0.05) at 95% confidence interval according to Tukey’s test.

When comparing the FA composition determined by the Codex Alimentarius to the Terminalia catappa kernel oils, major differences are found. However, palm kernel oil presented the closest FA profile to the CR and CA oils, mainly due to relatively similar amounts of palmitic and oleic acids, around 40%. Palm kernel oil also presented a significant proportion of linoleic acid, around 13%, which was two times lower than the CR and CA. These results corroborate the edibility of Terminalia catappa kernel oils from purple and yellow varieties.

The evaluation of the FA composition of vegetable oils can provide a vital classification of its lipids related to nutritional indices, mainly due to the presence of essential fatty acids, which can be used to correlate it to the prevention of cardiovascular diseases. The nutritional quality indices of T. catappa kernel oils from purple (CR) and yellow (CA) varieties are presented in Table 3. For comparison purposes the indices from other tropical fruit oilseeds, Caryocar villosum, Bactris gasipaes, and Oenocarpus bacaba, are displayed in the same table.

Nutritional quality indices of <italic>Terminalia catappa</italic> kernel oils from purple (CR) and yellow (CA) varieties and other tropical fruit oilseeds
T. catappa oils Other tropical fruit seed oils
CR CA Caryocar villosum 1 Bactris gasipaes 2 Oenocarpus bacaba 3
0.63 0.68 0.61 ND 0.43
0.50 0.54 0.38 1.10 0.30
1.16 1.27 0.75 2.04 0.67
0.75 0.79 2.58 0.84 3.32
2.75 3.05 ND ND ND

Data represent the calculated results from mean values (n = 3) of FA fractions, according to the equations previously presented. P/S - Polyunsaturated/saturated fatty acid ratio, AI - Atherogenicity index, TI - Thrombogenicity index. h/H - Hypocholesterolemic/hypercholesterolemic ratio, COX - calculated oxidation value. ND - non-determined. 1 Lorenzo et al. (2020); 2 Santos et al. (2020); 3 Pinto et al. (2018).

The ratio between polyunsaturated and saturated acids (P/S) is a relevant index of the nutritional quality of oils intended for human consumption, as a higher proportion of PUFAs may prevent the increase in body weight in high-fat diets. Nutritional regulations suggest a P/S ratio above 0.4, although this index cannot be taken into account alone for a healthy diet (Domínguez et al., 2016). the P/S value determined for the CR and CA oils were inferior to the value determined by Ladele et al. (2016), 0.84, and by Janporn et al. (2015), 1.4. However, when considering the recommend value of 0.4 by European legislations (Domínguez et al., 2016) the CR and CA still exhibited superior values.

The atherogenicity (AI) and thrombogenicity (TI) indices in human intake can linked to an increase in cardiovascular and other chronic non-transmissible diseases, when these values are not as low as possible (Santos et al., 2021; Ulbricht and Southgate, 1991). The AI and TI of the T. catappa kernel oils from purple (CR) and yellow (CA) varieties were higher than the values from Caryocar villosum and Oenocarpus bacaba oils determined in the works of Lorenzo et al. (2020) and Pinto et al. (2018), and lower than Bactris gasipaes oil (Santos et al., 2020). The replacement of animal fats in reformulated meat products by vegetable oils with lower AI and TI have demonstrated a significant improvement from a nutritional perspective (Domínguez et al., 2016). The h/H ratio of CR and CA oils were similar, and lower than the other tropical fruit oilseeds. A high value for this index in lipid intake may be advantageous to reducing low-density lipoproteins (LDL) in cholesterol fractions (Santos-Silva et al., 2002).

The calculated oxidation capacity value, COX, has a strong correlation with the proportion of PUFAs in lipid sources, and therefore, is expected to be higher in oils with high contents of PUFAs because they are more susceptible to oxidation. The COX value was lower in the CR oil than CA oil, which could be explained by the significant difference in the amount of linoleic acid between them, higher in CA than CR. The COX values for both samples, CR and CA, were lower than the indices of 6.6, 7.3, 6.5 and 7.8, as determined in oils from non-conventional sources, black cumin seeds (Nigella sativa), grape seeds (Vitis vinifera), tomato seeds (Lycopersicon esculentum) and wheat germ (Triticum vulgare), respectively (Hassanien et al., 2014). The CA and CR oils displayed lower COX values when compared to conventional oilseeds, such as linseed (12.6 - 13.9), sunflower (1.94 - 9.16), rapeseed (4.2 - 4.4), and camelina oils (8.7 - 9.4) (Symoniuk et al., 2018). Furthermore, these data corroborate the advantageous use of the T. catappa kernel oils from purple (CR) and yellow (CA) varieties for agro-industrial applications.

Triacylglycerol composition

The composition of triacylglycerides (TAGs) in CR and CA oils is displayed in Table 4. Both oils exhibited a quite similar proportion of triacylglycerols. The most frequently estimated triacylglycerols in CR were PLO, POO, POP, OLO, PLP, OLL and OOO, which represent 74.73% of the total. On the other hand, the predominant triacylglycerols in CA oil were almost the same, but slightly different with PLO, POO, POP, OLO, PLP, OLL and PLL representing 74.74% of the total. It was observed that the composition of TAGs was mainly composed of unsaturated acylglycerols, SU2 and U3, as can be seen in Table 4, which should be expected considering that the oleic and linoleic acids were most frequently in its FAs profile. Triglycerides are an important group of lipid sources for human nutrition. The TAG composition of CR and CA oils presented the predominance in ECN52 followed by ECN54, which can be linked to a large amount of long-chain triglycerides and, therefore, their inclusion in human intake can be helpful for preventing cardiovascular diseases.

Estimated percentage of triacylglycerol composition of <italic>Terminalia catappa</italic> kernel oil from purple (CR) and yellow (CA) varieties.
ECN Triacylglycerol Terminalia catappa oil % (normalized)
CR CA
C48:0 PPP 3.13 3.87
C50:0 SPP 1.75 1.98
C50:1 POP 11.80 11.77
C50:2 PLP 6.94 9.06
C52:1 SOP 4.41 4.02
C52:2 SLP 2.59 3.09
C52:2 POO 14.85 11.95
C52:3 PLO 17.47 18.39
C52:4 PLL 5.14 7.08
C54:2 SOO 2.77 2.04
C54:3 SLO 3.26 1.14
C54:3 OOO 6.22 4.04
C54:4 SLL 0.96 1.21
C54:4 OLO 10.98 9.34
C54:5 OLL 6.46 7.19
C54:6 LLL 1.27 1.84
Triacylglycerol classes %
S3 4.88 5.85
S2U 25.74 27.94
SU2 44.45 41.81
U3 24.93 22.41

ECN: equivalent carbon number. P - Palmitic acid, S - Stearic acid, O - Oleic acid, L - Linoleic acid.

S = saturated acylglycerol and U = unsaturated acylglycerol. Data represent the calculated results from mean values (n = 3) of FA fractions, according to the software PrOleos®.

 Analyses of bioactive compounds and antioxidant capacity

The bioactive compound analyses for anthocyanin, ascorbic acid and total polyphenol contents, and the antioxidant activity (TEAC assay), in the Terminalia catappa L. kernel oils from purple and yellow varieties are presented in Table 5. The analyses were used to investigate the presence of these bioactive compounds, and their potential antioxidant action as preserving agents in CR and CA oils. It was observed that the CR oil displayed higher values for bioactive compounds and antioxidant activity than CA oils (p < 0.05), which can be related to the difference in its variety.

Bioactive substances and antioxidant capacity of <italic>Terminalia catappa</italic> kernel oils from purple (CR) and yellow (CA) varieties.
Assays Samples
CR CA
Anthocyanin content (mg·100 g-1) 18.3 ± 1.5a 2.55 ± 1.03b
Ascorbic acid content (mg·100 g-1) 68.48 ± 2.02a 38.7 ± 1.5b
Total polyphenols content (mg·GAE g-1) 152.3 ± 2.4a 127.3 ± 3.0b
Antioxidant activity (μg·TE g-1) 38.6 ± 2.2a 31.1 ± 1.6b

Data represent the mean ± standard deviation of triplicate analyses (n = 3). Different superscript lowercase letters in the same line represent significant differences between samples (p < 0.05) at 95% confidence interval according to Tukey’s test.

In another work, the antioxidant activity of oils from T. cattapa was evaluated by the DPPH assay and its EC50 was found to be close to 7 mg·mL-1, indicating a potential antioxidant action (Ladele et al., 2016). Castelo-Branco, Torres (2012) investigated the antioxidant activity by TEAC assay of conventional oilseeds, such as soyabean, maize, sunflower, and canola, and found values close to 7.1, ~4.5, ~4.3 and ~5.3 mmol of Trolox eq·Kg-1 of oil, respectively.

A precise comparison of data from bioactive compounds and antioxidant activity in T. catappa oil was difficulted because of the scarce information available. Furthermore, different results found in the literature can be related to differences in protocol, sample preparation, solvents used, variations among species, harvest season, environmental conditions, and others.

Conclusions

The solvent extraction of Terminalia catappa L. kernel oils from purple (CR) and yellow (CA) varieties showed a good yield, above 54% lipids. These unconventional oils presented high-quality physical-chemical parameters, mainly observed by low levels of acidity and peroxides. Both oils, purple and yellow varieties, exhibited the predominance of unsaturated fatty acids (UFAs), with almost 63% oleic and 26% linoleic acids, which influenced its nutritional quality index values. These oils presented higher values for polyunsaturated and saturated acids ratio, which is relevant to human diets. The atherogenicity and thrombogenicity indices were higher in the T. catappa oils when compared to other tropical oilseeds. The calculated oxidation capacity values for both oils were lower than other non-conventional oil sources, even with a high proportion of PUFAs. The composition of triacylglycerols was mainly composed of unsaturated acylglycerols, which may be helpful for preventing cardiovascular diseases with their inclusion in human intake or in a potential use in formulated products with improvements in their nutritional profiles. Besides the nutritional quality properties, T. catappa oils from both varieties exhibited significative contents in anthocyanin, ascorbic acid and total polyphenol contents, and good antioxidant activity as determined by the TEAC assay. Thus, the presented results confirm the importance of T. catappa as a lipid source for human intake and to be used in the development of food products.

Acknowledgments

The authors acknowledge Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) Finance Code 001.

 References Abdulkadir AR 2015 In vitro antioxidant activity of ethanolic extract from Terminalia catappa (l.) leaves and fruits: Effect of fruit ripening Int. J. Sci. Res. 8 1244 1249 Agu CM Menkiti MC Nwabanne JT Onukwuli OD 2019 Comparative assessment of chemically modified Terminalia catappa L. kernel oil samples - A promising ecofriendly transformer fluid Ind. Crops Prod. 140 111727 10.1016/j.indcrop.2019.111727 Aliakbarian B Casazza AA Perego P 2011 Valorization of olive oil solid waste using high pressure-high temperature reactor Food Chem. 128 704 710 10.1016/j.foodchem.2011.03.092 Antoniosi Filho NR Mendes OL Lanças FM 1995 Computer prediction of triacylglycerol composition of vegetable oils by HRGC Chromatograph. 40 557 562 10.1007/BF02290268 AOCS - American Oil Chemist’s Society 2004 Official methods and recommended practices of the AOCS 5 Champaign, Illinois Ben EE Asuquo AE Owu DU 2019 Comparative effect of aspirin, meloxicam and Terminalia catappa leaf astanhola serum levels of some inflammatory markers in alloxan induced diabetic rats Asian J. Res. Biochem. 4 1 10 10.9734/ajrb/2019/v4i130058 Castelo-Branco VN Torres AG 2012 Generalized linear model describes determinants of total antioxidant capacity of refined vegetable oils Eur. J. Lipid Sci. Technol. 114 332 342 10.1002/ejlt.201100181 Chen Y Huang B He J Han L Zhan Y Wang Y 2011 In vitro and in vivo antioxidant effects of the ethanolic extract of Swertia chirayita J. Ethnopharmacol. 136 309 315 10.1016/j.jep.2011.04.058 Codex Alimentarius 2001 Codex standards for fats and oils Codex Stand 210 - 1999 FAO WHO Food Standards Second 2001 http://www.fao.org/3/y2774e/y2774e03.htm#bm3/ March 14, 2019 Cunha-Santos ECE Viganó J Neves DA Martínez J Godoy HT 2019 Vitamin C in camu-camu [Myrciaria dubia (H.B.K.) McVaugh]: evaluation of extraction and analytical methods Food Res. Int. 115 160 166 10.1016/j.foodres.2018.08.031 Domínguez R Agregán R Gonçalves A Lorenzo JM 2016 Effect of fat replacement by olive oil on the physico-chemical properties, fatty acids, cholesterol and tocopherol content of pâté Grasas Aceites 67 e133 10.3989/gya.0629152 Freitas MLF Chisté RC Polachini TC Sardella LACZ Aranha CPM Ribeiro APB Nicoletti VR 2018 Quality characteristics and thermal behavior of buriti (Mauritia flexuosa L.) oil Grasas Aceites 68 220 10.3989/gya.0557171 Ghafoor K Özcan MM AL-Juhaimi F Babiker EE Fadimu GJ 2019 Changes in quality, bioactive compounds, fatty acids, tocopherols, and phenolic composition in oven- and microwave-roasted poppy seeds and oil LWT - Food Sci. Technol 99 490 496 10.1016/j.lwt.2018.10.017 Hassanien MMM Abdel-Razek AG Rudzińska M Siger A Ratusz K Przybylski R 2014 Phytochemical contents and oxidative stability of oils from non-traditional sources Europ. J. Lipid Sci. Technol. 116 1563 1571 10.1002/ejlt.201300475 Huang Y-H Wu P-Y Wen K-C Lin C-Y Chiang H-M 2018 Protective effects and mechanisms of Terminalia catappa L. astanhola extract on hydrogen-peroxide-induced oxidative stress in human skin fibroblasts BMC Complementary Altern. Med. 18 266 275 10.1186/s12906-018-2308-4 ISO - International Organization for Standardization 2000 ISO 5509 2000 Animal and vegetable fats and oils - Preparation of methyl esters of fatty acids ISO Geneva, Switzerland Janporn S Ho C-T Chavasit V Pan M-H Chittrakorn S Ruttarattanamongkol K Weerawatanakorn M 2015 Physicochemical properties of Terminalia catappa seed oil as a novel dietary lipid source J. Food Drug Anal. 23 201 209 10.1016/j.jfda.2014.06.007 Jokić S Svilović S Vidović S 2015 Modelling the supercritical CO2 extraction kinetics of soybean oil Croat. J. Food Sci. Technol. 7 52 57 10.17508/CJFST.2015.7.2.05 Ladele B Kpoviessi S Ahissou H Gbenou J Kpadonou-Kpoviessi B Mignolet E Marie-France H Bero J Larondelle Y Leclercq JK Moudachirou M 2016 Chemical composition and nutritional properties of Terminalia catappa L. oil and kernels from Benin C. R. Chim. 19 876 883 10.1016/j.crci.2016.02.017 Lorenzo ND Santos OV Lannes SCS 2020 Fatty acid composition, cardiovascular functionality, thermogravimetric-differential, calorimetric and spectroscopic behavior of pequi oil (Caryocar villosum (Alb.) Pers.) Food Sci. Technol. 28 1 6 10.1590/1678-457X.0090 Menkiti MC Agu CM Udeigwe TK 2015 Extraction of oil from Terminalia catappa L.: Process parameter impacts, kinetics, and thermodynamics Ind. Crops Prod. 77 713 723 10.1016/j.indcrop.2015.08.019 Pinto RHH Sena C Santos OV Costa WA Rodrigues AMC Carvalho Junior RN 2018 Extraction of bacaba (Oenocarpus bacaba) oil with supercritical CO2: Global yield isotherms, fatty acid composition, functional quality, oxidative stability, spectroscopic profile and antioxidant activity Grasas Aceites 69 246e 1 8 10.3989/gya.0883171 Salawu AR Onyegbula AF Lawal IO Akande SA Oladipo AK 2018 Comparative study of the nutritional, phytochemical and mineral compositions of the nuts of Tropical Almond (Terminalia catappa) and Sweet Almond (Prunus amygdalus) Ruhuna J. Sci. 9 70 10.4038/rjs.v9i1.37 Santos OV Lorenzo ND Souza ALG Costa CEF Conceição LRV Lannes SCS Teixeira-Costa BE 2021 CO2 supercritical fluid extraction of pulp and nut oils from Terminalia catappa fruits: Thermogravimetric behavior, spectroscopic and fatty acid profiles Food Res. Int. 139 109814 10.1016/j.foodres.2020.109814 Santos OV Soares SD Dias PC Duarte SPA Santos MPL Nascimento FCA 2020 Chromatographic profile and bioactive compounds found in the composition of pupunha oil (Bactris gasipaes Kunth): implications for human health Rev. Nutr. 33 1 12 10.1590/1678-9805202033e190146 Santos-Silva J Bessa RJB Santos-Silva F 2002 Effect of genotype, feeding system and slaughter weight on the quality of light lambs Livest. Prod. Sci. 77 187 194 10.1016/S0301-6226(02)00059-3 Silva MSD Alves-Santos AM Santos IDD Wagner R Naves MMV Cordeiro MWS 2020a A new population of pequi (Caryocar spp.) developed by Brazilian indigenous people has agro-industrial and nutraceutical advantages Eur. Food Res. Technol. 246 1715 1724 Silva JCM Nicolau CL Cabral MRP Costa ER Stropa JM Silva CAA Scharf DR Simionatto EL Fiorucci AR Oliveira LCS Simionatto E 2020b Thermal and oxidative stabilities of binary blends of esters from soybean oil and non-edible oils (Aleurites moluccanus, Terminalia catappa, and Scheelea phalerata) Fuel. 262 116644 10.1016/j.fuel.2019.116644 Silva LMR Figueiredo EAT Ricardo NMPS Vieira IGP Figueiredo RW Brasil IM Gomes CL 2014 Quantification of bioactive compounds in pulps and by-products of tropical fruits from Brazil Food Chem. 143 398 404 10.1016/j.foodchem.2013.08.001 Singh SP Choudhary MR 2012 Indian Almond in Production technology of fruit crops in Wasteland Scientific Publishers Jodhpur, India Souza ALG Ferreira MCR Miranda LR Silvino RCAS Lorenzo ND Correa NCF Santos OV 2016 Aproveitamento nutricional e tecnológico dos frutos da Castanhola (Terminalia catappa Linn.)* Revista Pan-Amazônica de Saúde 7 23 29 10.5123/S2176-62232016000300003 Symoniuk E Ratusz K Ostrowska-Ligęza E Krygier K 2018 Impact of Selected Chemical Characteristics of Cold-Pressed Oils on their Oxidative Stability Determined Using the Rancimat and Pressure Differential Scanning Calorimetry Method Food Anal. Methods 11 1095 1104 10.1007/s12161-017-1081-1 Ulbricht TL Southgate DA 1991 Coronary heart disease: seven dietary factors The Lancet 338 985 992 10.1016/0140-6736(91)91846-M