GC-MS analysis of mango stem bark extracts (Mangifera indica L.), Haden variety. Possible contribution of volatile compounds to its health effects

1 Introduction

Mango stem bark extract (MSBE) has been developed as a bioactive ingredient for nutraceutical, cosmeceutical, and pharmaceutical formulations due to its antioxidant, anti-inflammatory, and analgesic effects [1]. The MSBE’s major component is a xanthone (mangiferin, 2-β-d-glucopyranosyl-1,3,6,7-tetrahydroxyl-9H-xanthen-9-one, hereafter MF), which has been intensively studied as a promising candidate to be developed for neurodegenerative diseases treatment [2], besides its use as antioxidant in food formulations against lifestyle disorders [3]. Our first work about the chemical composition of the MSBE led to the isolation of seven phenolic components: gallic acid and its methyl and propyl esters, MF, (+)-catechin, (−)-epicatechin, benzoic acid and its propyl ester, and 3,4-dihydroxybenzoic acid; four sugars: glucose, galactose, arabinose, and fructose; and 3 polyols: sorbitol, myoinositol, and xylitol [4]. All previous reported components from the MSBE were analyzed by high performance liquid chromatography with photodiode detection coupled to mass spectrometry (HPLC-DAD-MS), but no report has been published about its volatile components, which are usually analyzed by high resolution gas chromatography coupled to mass spectrometry (GC-MS). The composition of polyphenol-rich extracts from mango by-products (stem bark and branch tree) on two varieties (Haden and Tommy Atkins) has been compared, and we concluded that the Haden variety would be the best choice for a future exploitation of mango by-products for the production of polyphenol-rich extracts [5].

We report in this manuscript the MSBE volatiles’ composition from the Haden variety by GC-MS to contribute to its full chemical characterization. The possible influence of several volatile components on previous described MSBE health effects (antioxidant and anti-inflammatory) and their possible synergistic effects with nonvolatile components are discussed.

2 Materials and methods

3 Results

Figure 1 shows the chromatograms from extracts A to E, respectively. Ten compounds from 14 (71%) in extract A, nine compounds from 11 (82%) in extract B, 29 compounds from 53 (55%) in extract C, 20 compounds from 31 (65%) in extract D, and seven compounds from 10 (70%) in extract E could be identified by their mass spectra either by library search software or by comparison with mass spectra of pure standard compounds when available.

Figure 1

Gas chromatograms (FID) of mango stem bark extracts, Haden variety. (a) Fresh mango stem bark extract by Soxhlet extraction. (b) Industrial spray-dried mango stem bark extract (MSBE) by simultaneous steam distillation-solvent extraction. (c) Hexane eluate by preparative chromatography from extract A. (d) Hexane–ethyl acetate (1:1) eluate by preparative chromatography from extract A. (e) Ethyl acetate eluate by preparative chromatography from extract A.

Major components on extract A (Figure 1a) were peaks 1, 5, 6, 8, and 12, identified as β-elemene, β-selinene, 3,7(11)-selinadiene, bulnesol, and palmitic acid, respectively. Peaks 10 and 14 could be identified as a saturated hydrocarbon, with more than 17 carbons, and a gliceride from 9-octadecenoic acid, respectively, but fragmentation patterns were not conclusive. β-elemene (peak 1) and palmitic acid (peak 12) were identified by comparison of their chromatographic behavior and identical mass spectra of pure standards. β-selinene (peak 5), 3,7(11)-selinadiene (peak 6), and bulnesol (peak 8) matched library spectra with values of 0.99, 0.96, and 0.98, respectively, and had identical molecular ions as compared to published mass spectra data of pure compounds. Other identified minor components in extract A by comparison with authentic standards were trans-(−)-caryophyllene (peak 2), α-humulene (peak 3), n-hexadecane (peak 7), n-heptadecane (peak 9), and palmitic acid (peak 12).

Major components on extract B (SDE extraction, Figure 1b) were peaks 2, 3, 4 and 7, identified as β-elemene, β-selinene, 3,7(11)-selinadiene, and juniper camphor, respectively.

Peak 8 could be identified as a saturated aldehyde with more than 10 carbons, but mass spectrum was not conclusive. β-elemene (peak 2) was identified by comparison of its chromatographic behavior and identical mass spectrum of an authentic standard. β-selinene, 3,7(11)-selinadiene, and juniper camphor matched library spectra with values of 0.99, 0.96, and 0.99, respectively, and had identical molecular ions as compared to published mass spectra data of pure compounds. Other identified minor components in extract B by comparison with authentic standards were guaiol (peak 5), myristic acid (peak 9), palmitic acid (peak 10), and heptadecanenitrile (peak 11). Peak 6 was identified as α-eudesmol with a matched library spectrum of 0.94 and identical molecular ion as compared to published mass spectrum data of the pure compound.

Major components on extract C (nonpolar fraction, Figure 1c) were peaks 8, 10, 11, 19, and 29, identified as β-elemene, α-guaiene, (+)-aromandendrene, hinesol, and 1-octadecene, respectively. Peak 42 was identified as squalene, a common compound from stationary phase column bleeding. β-elemene (peak 8), (+)-aromandendrene (peak 11), hinesol (peak 19), and 1-octadecene (peak 29) were identified by comparison of their chromatographic behavior and identical mass spectra of pure standards. Peak 10 was identified as α-guaiene with a matched library spectrum of 0.95 and identical molecular ion as compared to published mass spectrum data of the pure compound. Other identified minor components in extract C by comparison with authentic standards were octanal (peak 1), dodecanal (peak 2), (+)-nootkatone (peak 27), 3-methyldibenzothiophene (peak 31), and β-amyrin (peak 49). Another 21 compounds were identified with matched library spectra values between 0.92 and 0.99 and had identical molecular ions as compared to published mass spectra data of pure compounds (see Table 1).

Major components on extract D (medium-polarity fraction, Figure 1d) were peaks 10, 15, 19, 23, and 31, identified as β-selinene, β-eudesmol, methyl linoleate, 9-methyl-(3β,5α)-androstan-3-ol, and γ-sitoesterol, respectively. Peaks 24–30 were identified as alkyl substituted phenols, but identification was not possible due to their low concentrations.

β-eudesmol (peak 15) and methyl linoleate (peak 19) were identified by comparison of their chromatographic behavior and identical mass spectra of pure standards. β-selinene (peak 10), 9-methyl-(3β,5α)-androstan-3-ol (peak 23), and γ-sitoesterol (peak 31) matched library spectra with values of 0.99, 0.96, and 0.98, respectively, and had identical molecular ions as compared to published mass spectra data of pure compounds. Other identified minor components in extract D by comparison with authentic standards were 6-methyl-3-heptanol (peak 1), phenol (peak 2), and palmitic acid (peak 16). Another 11 compounds were identified with matched library spectra values between 0.91 and 0.99 and had identical molecular ions as compared to published mass spectra data of pure compounds (see Table 1).

The major component on extract E (polar fraction, Figure 1e) was peak 8, identified as an oxygenated sesquiterpenoid (C₁₅H₂₆O), probably a naphtalenemethanol derivative, but fragmentation pattern was not conclusive. Hexanoic acid (peak 1) and β-eudesmol (peak 6) were identified by comparison of their chromatographic behavior and identical mass spectra of pure standards. β-chamigrene (peak 3), α-selinene (peak 5), 2,5-dihydroxy-α-methylphenetyl alcohol (peak 7), and N-phenyl-2-naphtaleneamine (peak 9) were identified with matched library spectra values of 0.96, 0.96, 0.91, and 0.92, respectively, and had identical molecular ions as compared to published mass spectra data of pure compounds. Summarizing, 64 compounds from 96 detected chromatographic peaks (67%) could be identified either by comparison with pure authentic standards (23 compounds, 24%) or by library mass spectra matches between 0.91 and 0.99 (41 compounds, 43%). Results are shown in Table 1. Relevant chemical structures of phenols (7 compounds), terpenoids (20 compounds), and steroids (9 compounds), regarding their possible biological significance, are shown in Figures 2–4, respectively.

Figure 2

Chemical structures of identified phenols: (1) phenol, (2) 3o-cathecol (1,2-benzenediol), (3) 1-(2-hydroxy-5-methylphenyl)ethanone, (4) 2,5-dihydroxy-α-methyl-phenetyl alcohol, (5) 3,4,5-trimethoxyphenol, (6) 3-octylphenol, (7) 3-pentadecyl-phenol.

Figure 3

Chemical structures of identified terpenoid compounds: (1) β-elemene, (2) α-elemene, (3) trans-(−)caryophyllene, (4) α-humulene, (5) α-guaiene, (6) β-chamigrene, (7) γ-selinene, (8) β-selinene, (9) α-selinene, (10) 3,7(11)-selinadiene, (11) (+) aromadendrene, (12) (+) nootkatone, (13) spathulenol, (14) ledol, (15) hinesol, (16) bulnesol, (17) α-eudesmol, (18) β-eudesmol, (19) guaiol, (20) juniper camphor.

Figure 4

Chemical structures of identified steroid compounds: (1) 9-methyl-(3β,5α)-androstane-3-ol, (2) 3β-campesterol, (3) 3β,5α-4,4-dimethylcholesta-8,14-dien-3-ol acetate, (4) stigmast-4-en-3-one, (5) γ-sitoesterol, (6) β-amyrin (olean-12-en-3β-ol), (7) D:C-friedoolean-3-one (multifluorenone), (8) 24-methylencycloartanol, (9) 3β-cycloartane-3,25-diol.

4 Discussion

Volatile compounds in fruits and vegetables usually have been analyzed in terms of their contribution to aroma [8] and flavor [9] properties of fresh or processed products, as part of their contribution to organoleptic properties in terms of product acceptance by the consumer. Fresh or processed fruits, fruit juices, fruit residues after industrial processing (peel and seeds), flowers, pollen, and roots and leaves (fresh or dried) are the most frequently studied parts for the determination of volatiles [10]. Volatile essential oil components from stem, or SB, have been studied in order to determine possible insect attractants or pheromones [11], insect repellents [12], and compounds with certain pharmacological activities [13].

The discoveries of quinine in the Chinchona bark in the XVII century, which subsequently led to chloroquine and hydroxychloroquine for malaria treatment, and of salicylic acid in the willow SB extract (Salix alba), the precursor of the worldwide known aspirin (acetylsalicylic acid) in the XIX century, are examples of nonvolatile blockbuster bioactive components that gave a sound contribution and discovery of new drugs from natural products chemistry. The possible contributions of volatiles from, i.e., willow SB extracts [14], different parts of pomegranate including SB [15], and cinnamon bark extract [16], as bioactive components have been studied. Our interest was to determine the presence of possible bioactive volatile components in the MSBE, which may contribute to its antioxidant, anti-inflammatory, and analgesic effects through a possible synergy mechanism [17]. The antioxidant, anti-inflammatory, analgesic, and immune-regulatory effects of the MSBE have been tested both in vitro and in vivo [1]. The concentrations at which MSBE exhibited its antioxidant effect were extremely low, no prooxidant effect were observed, and protection to oxidative damage was highly significant [18,19,20,21,22].

Haden mango SB extract components were mainly of nonpolar nature, and compounds found in higher relative amounts were sesquiterpene hydrocarbons as β-elemene, β-selinene, α-guaiene, and aromandendrene, in that order. A report has indicated the high anti-proliferative activity of β-elemene on glioma cells, and it was also an inducer of apoptosis in these cell lines [23]. It has shown to inhibit atherosclerotic lesions by reducing vascular oxidative stress and maintaining the endothelial function by improving plasma nitrite levels [24]. Several studies have shown that β-elemene may be a mediator in cancer prevention through an autoimmune mechanism [25]. Therefore, the possible contribution of β-elemene to the MSBE health effects is considerably high according to these previous reports.

Aromandendrene has been reported as a component of Eucaliptus sp. essential oils, and has shown to have synergistic properties with 1,8-cineole against antibiotic-resistant pathogens [26], but no other biological effect has been reported. β-selinene and α-guaiene have been found in several essential oils [27], and therefore, have no distinctive contributions to MSBE pharmacological effects. β-selinene-rich essential oils (between 37 and 57%) have shown to have strong reducing power as compared to gallic acid and catechin; good ability to chelate iron II; to moderate free radical scavenging activity; and have acceptable anti-inflammatory and antipyretic activities [28]. The presence of several terpenes from the eudesmane family, like β-selinene, in the MSBE volatiles composition, highly related to the antioxidant activity, must be considered in further studies in the attempt to correlate this health effect with eudesmane-type terpenes.

Major polar components on MSBE extracts were juniper camphor, hinesol, β-eudesmol, 9-methyl-(3β,5α)-androstan-3-ol, γ-sitosterol, and 2,5-dihydroxy-methylphenetylalcohol. Although these oxygenated compounds were determined in relatively low amounts, as compared to sesquiterpene hydrocarbons, they have higher water solubility, and therefore, higher bioavailability in terms of their possible contribution to MSBE pharmacological effects. These components are commonly found in essential oils from several plants with antioxidant [28] or antimicrobial [29] effects. Hinesol has shown to inhibit H⁺, K⁺-ATPase [30], and this result may explain the observed benefits of MSBE on gastric disorders. It also enhanced the inhibitory effect of omeprazole on the hydrogen pump. On the other hand, β-eudesmol stimulates an increase in appetite through Transient Receptor Potential Ankyrin (TRPA1), and therefore body weight gain [31]. Again, the presence of eudesmane-type sesquiterpenic alcohols adds a new insight into their possible contribution to the MSBE antioxidant effect.

We conducted the analysis on an industrial MSBE (extract B), which has been used as an antioxidant bioactive ingredient on nutritional supplement formulations (tablets and capsules), and a cosmeceutical cream [7], in order to determine differences on volatiles composition as compared to fresh mango SB (extract A). Both extracts had similar content of β-elemene, β-selinene, and a naphthalene derivative (eudesmane type), but it could not be determined that juniper camphor was present in extract B, which was not present in extract A. Therefore, it might be assumed that these three nonpolar components would contribute to synergize the antioxidant and anti-inflammatory effects of other less volatile components of MSBE bioactive ingredients, like polyphenols and flavonoids, in commercial formulations. Juniper camphor (eudesmane family) is a common available component from plant extracts [32], but its possible contribution to biological effects of essential oils or plant extracts is not clear [33]. It has been found as the main component of Pulicaris sp. essential oils, which have shown high antioxidant activity [34], but reports about biological effects as pure compound are not available.

Studies about mango volatiles have shown that their composition may differ significantly among varieties [35]. The main research focus on mango volatiles has been to identify the major contributing compounds to fruit aroma and flavor [36,37]. The influence of germplasm on volatiles composition of mango fruit cultivated in seven countries has been studied [38]. A study on the volatile composition of the gum-resin exudated by the bark trunk of mango (non-specified variety) showed the presence of selinenes (α- and β-) as the major components, followed by β-caryophyllene, β-elemene, and β-chamigrene [39]. Our findings are consistent with these results in terms of sesquiterpene hydrocarbons, but oxygenated components were not found in gum-resin, probably due to the sampling technique (headspace). Nevertheless, reports about composition on the MSBE for any variety, and its possible health effects, are only few.

We reported previously 14 components, mainly nonvolatile polyphenols, sugars, and polyols, form an industrial MSBE [4] by HPLC, MS, and NMR techniques, and eight main minerals: Na, K, Ca, Mg, Mn, Cu, Zn, and Se by ICP-MS [40]. We report now a list of 64 volatile components in the MSBE, identified by GC-MS, which will give a sound basis in the attempts to correlate observed MSBE pharmacological effects to its chemical composition and their possible synergy effects with nonvolatile components.

5 Conclusions

The analysis of MSBE, Haden variety, by GC-MS is reported for the first time, which gives an approach in order to understand the possible synergistic effect of volatile compounds on its antioxidant, analgesic, and anti-inflammatory effects. The identification of relevant bioactive components in MSBE extracts, mainly terpenes and sesquiterpenes from the eudesmane family, will contribute to correlate their chemical composition to previous reported pharmacological effects.