Ceiba speciosa (A. St.-Hil.) Seeds Oil: Fatty Acids Profiling by GC-MS and NMR and Bioactivity.

This study aimed to evaluate the chemical composition by gas chromatography-mass spectrometry (GC-MS) and Nuclear Magnetic Resonance (NMR) analyses, the antioxidant activities evaluated by different in vitro assays namely 2,2-diphenyl-1-picrylhydrazyl (DPPH), 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid) (ABTS), Ferric Reducing Ability Power (FRAP), and β-carotene bleaching tests, and the inhibitory effects of enzymes linked to obesity (lipase, α-amylase, and α-glucosidase) of fixed seed oil of Ceiba speciosa (A. St.-Hil.). Fourteen compounds were identified. Linoleic acid (28.22%) was the most abundant followed by palmitic acid (19.56%). Malvalic acid (16.15%), sterculic acid (11.11%), and dihydrosterculic acid (2.74%) were also detected. C. speciosa fixed oil exerted a promising ABTS radicals scavenging activity with an IC50 value of 10.21 µg/mL, whereas an IC50 of 77.44 µg/mL against DPPH+ radicals was found. C. speciosa fixed oil inhibited lipase with an IC50 value of 127.57 µg/mL. The present investigation confirmed the functional properties of C. speciosa fixed oil, and proposes its use as valuable source of bioactive constituents.


Introduction
Metabolic syndrome is a multiple factor syndrome in which there is a co-existence of several risk factors such as hyperglycaemia, dyslipidaemia, and hypertension. This syndrome is frequently linked to obesity [1]. In fact, a dysregulated production of adipokines by adipocytes contributes to insulin resistance and atherosclerosis with an increase risk in the development of thrombosis [2,3]. Epidemiological studies evidence that worldwide the number of obese people is approximately 2.1 billion. This number continues to rise. Moreover, it is estimated that obesity and linked pathology kill over 3 million people each year [4]. Oxidative stress plays critical roles in the pathogenesis of both diabetes and obesity. In diabetic patients, oxidative stress impairs glucose uptake in both muscle and fat tissue, as well as reduces insulin secretion by β cells [5]. Table 1 reports the main fatty acids identified by GC-MS analysis in C. speciosa seeds n-hexane extract. A total of 14 main compounds were identified ( Figure 1). Linoleic acid (28.22%) was the most abundant fatty acid, followed by palmitic acid (19.56%). The three cyclopropane fatty acids, identified by comparison with published mass spectra [29] such as malvalic acid, sterculic acid and dihydrosterculic acid accounted for 16.15, 11.11, and 2.74%, respectively. Bohannon and Kleiman [28] reported percentage of malvalic acid and sterculic acid of 12.4 and 10.0%, respectively. Data are reported as mean ± standard deviation (n= 3). a Identified al methyl esters b Identification by comparison with published data [29].

Chemical Profiling
Molecules 2020, 25, x FOR PEER REVIEW 3 of 12 Table 1 reports the main fatty acids identified by GC-MS analysis in C. speciosa seeds n-hexane extract. A total of 14 main compounds were identified ( Figure 1). Linoleic acid (28.22%) was the most abundant fatty acid, followed by palmitic acid (19.56%). The three cyclopropane fatty acids, identified by comparison with published mass spectra [29] such as malvalic acid, sterculic acid and dihydrosterculic acid accounted for 16.15, 11.11, and 2.74%, respectively. Bohannon and Kleiman [28] reported percentage of malvalic acid and sterculic acid of 12.4 and 10.0%, respectively.
Previously, lower percentages of these cyclopropane fatty acids (9.0 and 6.0% for malvalic and sterculic acids, respectively) were found by Petronici et al. [19]. The chemical analysis of Ceiba pentandra seed oil from Malaysia showed the presence of linoleic acid (38.82%), palmitic acid (24.31%), and oleic acid (21.88%) as the most abundant fatty acids [30]. Malvalic and sterculic acids were also identified with percentage of 7.18 and 2.96%, respectively [30]. Differently, Kiran and Rao [31] reported a lower percentage of linoleic acid for the seed oil of C. pentadra from India. Data are reported as mean ± standard deviation (n= 3). a Identified al methyl esters b Identification by comparison with published data [29].
The GC-MS analysis of the trans-methylated fixed oil showed the occurrence of eleven fatty acids methyl esters and three other compounds that could not be identified through our library.  Previously, lower percentages of these cyclopropane fatty acids (9.0 and 6.0% for malvalic and sterculic acids, respectively) were found by Petronici et al. [19]. The chemical analysis of Ceiba pentandra seed oil from Malaysia showed the presence of linoleic acid (38.82%), palmitic acid (24.31%), and oleic acid (21.88%) as the most abundant fatty acids [30]. Malvalic and sterculic acids were also identified with percentage of 7.18 and 2.96%, respectively [30]. Differently, Kiran and Rao [31] reported a lower percentage of linoleic acid for the seed oil of C. pentadra from India.
The GC-MS analysis of the trans-methylated fixed oil showed the occurrence of eleven fatty acids methyl esters and three other compounds that could not be identified through our library.
Therefore, on the basis of previous works carried out on the genus Ceiba [28,30], which revealed the presence of cyclopropanic fatty acids, 1D and 2D-NMR techniques were employed to search for the characteristic signals of these acids. Figure 2 shows signals, in the 1 H-NMR spectrum, characteristic of dihydrosterculic, sterculic, and malvalic methyl ester. The signal between −0.33 (q) can be attributed to one proton of the methylene moiety in the cyclopropanic ring of methyl dihydrosterculate [32]; while the singlet at 0.77 ppm is generated by cyclopropenic methylene protons, such as those present on sterculic and malvalic methyl esters [33,34]. Therefore, on the basis of previous works carried out on the genus Ceiba [28,30], which revealed the presence of cyclopropanic fatty acids, 1D and 2D-NMR techniques were employed to search for the characteristic signals of these acids. Figure 2 shows signals, in the 1 H-NMR spectrum, characteristic of dihydrosterculic, sterculic, and malvalic methyl ester. The signal between −0.33 (q) can be attributed to one proton of the methylene moiety in the cyclopropanic ring of methyl dihydrosterculate [32]; while the singlet at 0.77 ppm is generated by cyclopropenic methylene protons, such as those present on sterculic and malvalic methyl esters [33,34]. However, due to the overlapping of signals in the 1 H-NMR spectrum, 13 C-NMR spectrum and heteronuclear correlation (HSQC) (Figure 3 and Figure 4) were registered in order to confirm the presence of the cyclopropanoid acids. The 13 C NMR spectrum showed 3 typical carbon signals at 10.89 ppm for the methylene and at 15.71 and 15.75 ppm for the methine carbons of the cyclopropane ring, respectively, of methyl dihydrosterculate. The methylene protons (δHa = 0.57, δHb = −0. 34) showed HSQC correlation with a secondary carbon C3 at δC3 = 10.89; a single cross correlation peak was also visible for the methine protons (δHc = 0.63) and the carbons C1 and C2 at 15.71 and at 15.75 ppm, respectively. These data strongly suggested the occurrence of methyl dihydrosterculate. In turn, for the sterculic and malvalic methyl ester, the peak of the cyclopropenic methylene protons at 0.77 ppm correlates with 13 C NMR signal at 7.35 ppm (C-3).
Furthermore, the signals at 109.19 ppm and 109.42 ppm, and 109.11 ppm and 109.50 ppm were in perfect agreement [34] with those reported for sterculic and malvalic methyl ester, respectively. Based on these considerations, the mass-spectra and the retention indexes of the three GC unidentified compounds were searched in the literature [29] and a perfect match with the cyclopropanoid fatty acid methyl esters was found. However, due to the overlapping of signals in the 1 H-NMR spectrum, 13 C-NMR spectrum and heteronuclear correlation (HSQC) (Figures 3 and 4) were registered in order to confirm the presence of the cyclopropanoid acids. The 13 C NMR spectrum showed 3 typical carbon signals at 10.89 ppm for the methylene and at 15.71 and 15.75 ppm for the methine carbons of the cyclopropane ring, respectively, of methyl dihydrosterculate. The methylene protons (δ Ha = 0.57, δ Hb = −0.34) showed HSQC correlation with a secondary carbon C3 at δ C3 = 10.89; a single cross correlation peak was also visible for the methine protons (δ Hc = 0.63) and the carbons C1 and C2 at 15.71 and at 15.75 ppm, respectively. These data strongly suggested the occurrence of methyl dihydrosterculate. In turn, for the sterculic and malvalic methyl ester, the peak of the cyclopropenic methylene protons at 0.77 ppm correlates with 13 C NMR signal at 7.35 ppm (C-3).
Furthermore, the signals at 109.19 ppm and 109.42 ppm, and 109.11 ppm and 109.50 ppm were in perfect agreement [34] with those reported for sterculic and malvalic methyl ester, respectively. Based on these considerations, the mass-spectra and the retention indexes of the three GC unidentified compounds were searched in the literature [29] and a perfect match with the cyclopropanoid fatty acid methyl esters was found.

Antioxidant Activity
The antioxidant activities of C. speciosa seed fixed oil were investigated by using two radical scavenging assays, namely DPPH and ABTS. The protection of lipid peroxidation (β-carotene bleaching test) as well as the effect on iron, one of the main involved ion in the oxidation process, was studied. Data are reported in Table 2.

Antioxidant Activity
The antioxidant activities of C. speciosa seed fixed oil were investigated by using two radical scavenging assays, namely DPPH and ABTS. The protection of lipid peroxidation (β-carotene bleaching test) as well as the effect on iron, one of the main involved ion in the oxidation process, was studied. Data are reported in Table 2.
Fixed oil showed a concentration-dependent antioxidant activity in both radical scavenging test. In particular, C. speciosa fixed oil exerted a promising ABTS radicals scavenging activity with an IC 50 value of 10.21 µg/mL, whereas it had an IC 50 of 77.44 µg/mL against DPPH + radical. These results may be justified taking into consideration the different mechanisms of action of the two radicals scavenging assays.
A lower effect was recorded in the FRAP test in which the FRAP value was 3-times lower than that the positive control BHT. Percentage of 37.36 and 36.98% were obtained in β-carotene bleaching test at oil concentration of 100 µg/mL after 30 and 60 min of incubations, respectively. Data are expressed as means ± S.D. (n = 3). Differences within and between groups were evaluated by one-way ANOVA followed by a multi-comparison Dunnett's test α = 0.05): **** p< 0.0001, ** p < 0.01, compared with the positive controls.
To our knowledge, this is the first report on the antioxidant activity of fixed oil. Previously, Dörr et al. [35] revealed that C. speciosa aqueous steam bark extract exerted a promising DPPH radical scavenging potential. A promising DPPH radical scavenging effect was observed, also by Krüger Cardoso Malheiros et al. [36], who found percentage of 85.13 and 88.95% at concentration of 50 µg/mL for raw aqueous bark extract and ethanol extract, respectively. Both extracts are rich in kaempferol, cholorogenic acid and caffeic acid. Different extracts from C. speciosa leaves and stem were studied for their antioxidant activity [25]. Generally, the stem was a little more active than leaves and fruits with percentage of inhibition of 81.2, 78.87, and 77.5% for stem, leaves, and fruits ethanol extract, respectively. More recently, Refaat et al. [27] compared the DPPH radicals scavenging ability of different extracts from C. speciosa and C. chodatii. Ethyl acetate extracts had the most active fractions followed by aqueous extract. Comparing different investigated parts, the following trend: flowers > fruits > leaves > seeds was found.

Inhibition of Enzymes Linked to Obesity
The search for new agents useful for the prevention of obesity and related pathologies is a topic of great interest for the scientific community. C. speciosa fixed oil was investigated for its potential anti-obesity and related disorder prevention effect. Data are reported in Table 3.
A concentration-dependent manner should be observed for all investigated samples independently by the enzyme assays used. IC 50 values of 135.69 and 158.22 µg/mL were recorded against α-amylase and α-glucosidase, respectively, whereas IC 50 value of 127.57 µg/mL was found against lipase enzyme. The in vivo anti-hyperglycaemic effect of C. insignis was observed by El-Alfy et al. [37]. Dried leaves' total and aqueous extracts as well as ethyl acetate fraction administered intraperitoneally (150 mg/kg) exerted a promising anti-hyperglycaemic by decreasing blood glucose level effect in alloxan-induced diabetic rats as compared with the positive control metformin.
The determination of the median lethal dose (LD 50 ) of the total and the aqueous extracts and the successive fractions revealed that this plant is safe. The effect of C. pentandra on obese wistar albino rats was observed by Patil et al. [38]. Oral administration of 125 mg/kg of leaves total ethanol extract prevented the increase of body weight, as demonstrated by the reduction of body max index Molecules 2020, 25, 1037 8 of 12 (0.16 g/cm 2 ) as compared to the cafeteria control group without any effect on appetite. Moreover, administration of C. pentandra extract did not decrease total cholesterol content, LDL or HDL levels. Therefore, it is possible that its action is mediated by preventing the breakdown of dietary fat in the gastrointestinal tract. The acute toxicity study revealed that administration of C. pentandra extract was safe up to 5000 mg/kg. More recently, Refaat et al. [27] demonstrated that both C. speciosa and C. chodati total ethanol extracts of different plant portions at 5 µg/ mL reduced the lipid droplets in 3T3-L1 preadipocytes. However, at higher doses, an induction on adipogenesis was observed. Data are expressed as means ± S.D. (n = 3). Acarbose used as positive control in α-amylase and α-glucosidase tests. Orlistat used as positive control in the lipase test. Differences within and between groups were evaluated by one-way ANOVA, followed by a multicomparison Dunnett's test (α = 0.05): **** p < 0.0001 compared with the positive control.

Plant Materials
The fruits of C. speciosa were harvested in May 2019 from different plants growing in the campus of Palermo University, Parco d'Orleans, Palermo (Sicily, Italy) (38 • 06 10" N, 13 • 20 52" E) at 47 m. a.s.l. In particular, the gathering of fruits was carried out on the campus by collecting them randomly from 20 healthy plants. The collected fruits were placed in a laboratory and air dried for about a month. The seeds were manually drawn out from the fruits and used for the analysis.

Extraction Procedure
Seeds (50 g) were removed from the fruits, then reduced to a fine powder using a blender type A11 basic and extracted twice with n-hexane (400 mL) under stirring at room temperature for 72 h. The solvent was evaporated at 40 • C using a Buchi rotavapor R-200 (Germany) to give 7.33 g of oil (yield 14.6%) [39]. The resulting oil was kept away from light and at low temperature.

Gas Chromatography-Mass Spectrometry (GC-MS) Analyses
Fixed oil (100 mg) was subjected to basic transmethylation using potassium hydroxide in methanol. To a solution of the oil (100 mg in 2 mL of hexane), 0.2 mL of 2 M methanolic KOH was added and allowed to stir for 2 min or at 30 • C [40]. An aliquot of the upper hexane layer was directly taken and analysed by gas chromatography associated with mass spectrometry (GC-MS), using a Hewlett-Packard 6890 gas chromatograph (Agilent, Milan, Italy) equipped with a non-polar HP-5 capillary column (30 m × 0.25 mm, 0.25 µm), associated with a Hewlett-Packard 5973 mass spectrometer completed by Hewlett Packard Chemstation data system (Agilent, Milan, Italy). The ionization of the sample constituents was performed in electronic impact (EI, 70 eV). The analyses were carried out by using the following temperature schedule: isotherm at 50 • C for 5 min, temperature increase from 60 to 250 • C of 14 • C/min, and finally isotherm at 250 • C for 10 min. Helium is used as a carrier gas. Compounds identification corresponding to methyl esters (FAMEs) was based on the comparison of the mass spectral data with the Wiley 128 library and referring to the spectral data of a standard mixture of FAMEs. The compounds' relative concentrations were calculated based on peak areas without using correction factors.
The chemical shifts are reported relative to the chloroform residual peak (7.27 ppm for 1 H NMR).

Radical Scavenging Activity Assays
The 2,2-diphenyl-1-picrylhydrazyl (DPPH) radicals scavenging activity was performed as previously reported [41]. Absorbance modifications as a consequence of DPPH radical reaction was measured at 517 nm. 2,2 -Azino-bis(3-ethylbenzothiazoline-6-sulphonic acid) (ABTS) radical scavenging ability was measured following the protocol previously described [42]. ABTS scavenging activity (%) was calculated as follows: [(A -A)/A ] × 100, where A is the absorbance of the control reaction and A is the absorbance in the presence of the extract. Ascorbic acid was employed as a positive control in both assays.

β-Carotene Bleaching Test
The β-Carotene Bleaching Test measured the ability of the extract to protect lipid substrate from peroxidation. The test was applied following the procedure described by Loizzo et al. [41]. In brief, extract at different concentrations, β-carotene solution, linoleic acid and 100% Tween 20 were mixed. The absorbance was measured at 470 nm against a blank at t = 0 and after 30 and 60 min of incubation. Propyl gallate was used as a positive control.

Ferric Reducing Ability Power (FRAP) Assay
The FRAP assay was done following the previously described procedure [43]. The FRAP value represents the ratio between the slope of the linear plot for reducing Fe 3+ -TPTZ reagent by different extract compared to the slope of the plot for FeSO 4 .

Pancreatic Lipase Inhibitory Activity
The pancreatic lipase inhibitory activity was investigated following the procedure of El-shiekh with some modifications [44]. The enzyme at concentration of 1 mg/mL was measured using 4-nitrophenyl octanoate as a substrate, 5 mmol/L in dimethylsulfoxide solution. Extracts at different concentrations were left to react with enzyme, substrate, and Tris-HCl buffer (pH 8.5) at 37 • C for 30 min. The absorbance was measured (405 nm). Orlistat was used as a positive control.

Carbohydrates-Hydrolysing Enzymes Inhibitory Activity
The carbohydrates-hydrolysing enzymes' inhibitory activity was investigated using α-amylase and α-glucosidase as enzymes. The procedure is previously described [41]. Results are expressed as IC 50 values and acarbose was used as positive control in both assays.

Statistical Analysis
The concentration-response curve and the inhibitory concentration 50% (IC 50 ) was calculated by using Prism GraphPad Prism version 4.0 for Windows, GraphPad Software (San Diego, CA, USA). The same software was used to perform one-way ANOVA followed by a multicomparison Dunnett's test (α = 0.05): **** p < 0.0001, *** p < 0.001 compared with the positive controls.

Conclusions
The present study assessed the chemical profile, α-amylase, α-glucosidase, and lipase inhibitory activity, and the antioxidant properties of C. speciosa fixed oil obtained from plant domesticated in Palermo (Sicily). Linoleic acid was the most abundant fatty acid followed by palmitic acid. Three cyclopropane fatty acids, namely malvalic, sterculic, and dihydrosterculic, are also identified and their concentration is in line with those reported for other Ceiba species. The fixed oil exerts both hypoglycaemic and anti-obesity effects and for this reason data obtained from this research activity could help to support the use of the edible oil of a local plant due to its antioxidant properties and for the prevention of obesity and related disorders.