Objective: α-Mangostin (1), a xanthone predominantly abundant in Garcinia mangostana, exhibits diverse pharmacological activities. As reported in our previous study, it also shows inhibitory activities against neuraminidase of the influenza virus. In the present study, we investigated α-mangostin derivatives produced by γ-irradiation to identify more effective neuraminidase inhibitors. Methods: α-Mangostin (1) in methanol (2.439 mM) was treated with 200 Gy of gamma (60Co) irradiation (10 kGy/h). Five radiolysis products were isolated using combined chromatographic techniques, and their structures were identified using one- and two-dimensional nuclear magnetic resonance and high-resolution electrospray ionization mass spectroscopy. The neuraminidase inhibitory activities of all compounds were evaluated and compared. Results: Five derivatives (2−6) including two new compounds, 13,6-trihydroxy-2-(2-hydroxyethyl)-7-methoxy-8-(3-methylbut-2-enyl)-9H-xanthen-9-one (5) and 13,6-trihydroxy-2-(4-hydroxy-2-methoxy-3,3-dimethylbutyl)-7-methoxy-8-(3-methylbut-2-enyl)-9H-xanthen-9-one (6), were obtained by γ-irradiating α-mangostin dissolved in methanol. All derivatives exhibited IC50 values of 7.84–44.66 μM against neuraminidase. Mangostanin (4) exhibited an IC50 of 7.84 μM, which is 1.6- and 1.8-fold more potent than that of 1 and quercetin (positive control), respectively. In addition, all compounds exhibited competitive inhibition. Conclusion: Our results suggest that γ-irradiation is an effective method for structural modification of α-mangostin and it derivative, mangostanin is potential neuraminidase inhibitor.
Gamma (γ) irradiation is used to preserve food through the destruction of microorganisms by reactive species and free radicals.1 However, various scientific approaches have been applied to modify natural compounds using γ-irradiation for generating novel derivatives and for improving their bioactivity or pharmacokinetic properties.2 The molecules in a given solution are degraded due to indirect redox reactions caused by the reactive species generated by the radiolysis of the solvent. In addition, their chemical and biological modification may lead to the development of new drugs with improved efficacy.3 Therefore, this study aimed to apply ionizing radiation technology to α-mangostin for the production of its derivatives with enhanced biological potency.
α-Mangostin, a prenylated xanthone, is mainly isolated from the pericarp of mangosteen, the fruit of Garcinia mangostana.4–6 Previous studies have reported that α-mangostin exhibits various pharmacological properties such as anticancer,7,8 antioxidant,7 anti-inflammatory,7,8 anti-metabolic syndrome (eg, obesity, diabetes, and cardiovascular disease),7,9 anti-neurodegenerative,10,11 antimicrobial, and antiviral activities.6,12,13 We previously isolated xanthones from the seedcase of G. mangostana for the first time in the literature and demonstrated their inhibitory activity against neuraminidase, which plays a significant role in the infectivity of the influenza virus.6 Subsequently, the extent of the antiviral activity of α-mangostin has been explored. α-Mangostin inhibits Chikungunya virus infection and replication in vivo and in silico.14 Moreover, it demonstrates an inhibitory effect on dengue virus (DENV) infection and viral production and reduces cytokines and chemokines.15 It also suppresses DENV serotype-2 production at several stages of the replication cycle by interacting with multiple DENV protein targets.16 Therefore, α-mangostin or its derivatives may be potential therapeutic agents against several viral infections in human beings. However, no previous report on the structural modification of α-mangostin using γ-irradiation and the evaluation of neuraminidase inhibitory activities of its derivatives.
Results and Discussion
γ-Irradiation of α-Mangostin in Methanol
γ-Radiolysis of 2.439 mM of α-mangostin in methanol was performed using a dose of 200 kGy. The high-performance liquid chromatogram (HPLC) of the γ-irradiated α-mangostin solution is shown in Figure 1. The degradation of α-mangostin resulted in the formation of several compounds, and the radiochemical yield of each compound increased as the initial concentration of α-mangostin decreased. This phenomenon is consistent with the classical radiolysis process, in which the solute reacts with the radiolytic species due to the solvent.2 During the liquid chromatography of the α-mangostin sample irradiated above 200 kGy, the intensity of the correcting peak of α-mangostin did not decrease. For samples irradiated below 200 kGy, the number and intensity of peaks corresponding to radiolysis products were very weak. Therefore, our study suggests that the optimal dose to produce radiolysis products of α-mangostin is 200 Gy. However, in term of reproducibility of radiolysis production, further validation studies on the stable production of radiolysis products are required.
HPLC chromatograms (at 280 nm) of (A) α-mangostin and (B) γ-irradiated α-mangostin in a methanol solution (2.439 M) at a dose of 200 kGy.
Identification of Five Radiolysis Products 2–6
The structures of five radiolysis products 2–6, formed during the degradation of α-mangostin (1), were identified via the analysis of 1H, 13C, 1H-1H COSY, 1H-13C heteronuclear single quantum coherence (HSQC), and 1H-13C heteronuclear multiple bond correlation (HMBC) nuclear magnetic resonance (NMR) data, as well as high-resolution electron ionization mass spectroscopy (HR-ESI-MS) data (Figure 2). Compounds 2, 3, and 4 were identified as 3-isomangostin,5 11-hydroxy-1-isomangostin,17 and mangostanin,7 respectively, by comparing their spectroscopic data with literature values.
Chemical structures of compounds 1-6.
Compound 5 was obtained as a yellowish powder and exhibited a molecular ion at m/z 386.1368 [M]+ in HR-ESI-MS, corresponding to a molecular formula of C21H22O7. 1H and 13C spectra of 5 were similar to those of α-mangostin (1),18 except for the presence of 2-hydroxyethyl signals at δH 2.89 (2H, dd, J = 14.9, 7.5 Hz, H-10) and 3.66 (2H, t, J = 14.9, 7.7 Hz, H-11), as well as δC 25.4 (C-10) and 60.7 (C-11), instead of signals for one of the two sets of 3-methylbut-2-enyl groups in 1 (Tables 1 and 2). This 2-hydroxyethyl moiety was located at C-2, as evidenced by the 1H-13C HMBC correlations of H-10/C-1, C-2, C-3 and H-112/C-2, C-10 (Figure 3). In addition, the position of the 3-methylbut-2-enyl group was assigned to C-2 as per the 1H-13C HMBC correlations of H-13/C-7, C-8 and H-14/C-8. Further detailed analysis of the 1H-1H COSY, 1H-13C HSQC, and 1H-13C HMBC NMR data (Figure 2) enabled unambiguous assignments for all of 1H and 13C NMR signals of 5. Thus, compound 5 was determined as 13,6-trihydroxy-2-(2-hydroxyethyl)-7-methoxy-8-(3-methylbut-2-enyl)-9H-xanthen-9-one, which is a previously unknown compound.
Key 1H-1H COSY and 1H-13C HMBC correlations of 5 and 6.
1H NMR (500 MHz) Spectroscopic Data of Compounds 1–6 in CD3OD [δ in ppm, Mult. (J in Hz)].
H No.
1
2
3
4
5
6
4
6.25, s
6.24, s
6.30, s
6.24, s,
6.24, s
6.24, s
5
6.76, s
6.67, s
6.64, s
6.67, s
6.69, s
6.69, s
10
3.42, d (6.3)
2.67, m
2.56, dd, (18.0, 8.2) 2.91, dd (18.0, 5.6)
2.67, m
2.89, dd (14.9, 7.4)
2.81, m
11
5.26, t (6.3)
1.64, m
3.79, m
4.86, m
3.66, dd (14.9, 7.7)
3.41, m
12
4.06, d (6.6)
13
1.82, s
1.25, s
1.35, s
1.25, s
5.21, t (6.6)
3.52, m
14
1.68, s
1.25, s
1.47, s
1.25, s
0.97, s
15
4.06, d (5.7)
4.07, d (6.6)
4.05, d (5.8)
4.07, d (6.6)
1.64, s
0.93, s
16
5.26, t (5.7)
5.22, m
5.30, t (5.8)
5.22, m
1.81, s
4.07, d (6.5)
17
5.21, t (6.5)
18
1.75, s
1.65, s
1.83, s
1.67, s
19
1.82, s
1.81, s
2.02, s
1.81, s
1.66, s
20
1.82, s
OCH3-7
3.79, s
3.74, s
3.75, s
3.74, s
3.74, s
3.74, s
OCH3-11
3.17, s
13C NMR (125 MHz) Spectroscopic Data of Compounds 1–6.
C No.
1a
2b
3a
4b
5b
6b
1
160.6
160.3
156.9
160.3
161.0
160.9
2
108.8
110.8
1039
110.9
106.8
108.6
3
161.7
162.4
160.6
162.4
162.9
161.8
4
93.3
91.8
93.1
87.8
91.9
92.0
4a
155.0
156.8
160.6
156.8
155.4
155.2
5
101.7
101.5
100.9
101.5
101.5
101.4
5a
155.8
155.4
154.7
155.4
156.9
156.4
6
154.7
154.8
154.2
154.8
155.3
155.4
7
142.7
143.6
144.2
143.6
143.6
143.5
8
137.1
137.1
136.8
137.1
137.2
137.2
8a
112.2
110.9
113.5
110.8
110.8
110.8
9
182.1
182.4
177.5
182.4
181.8
181.8
9a
103.6
102.4
109.4
102.4
102.4
102.3
10
21.5
27.6
25.7
26.1
25.4
23.2
11
121.7
42.9
68.2
91.6
60.7
85.9
12
135.3
70.5
78.1
71.1
25.8
40.3
13
25.9
25.8
19.3
24.7
123.8
68.8
14
25.9
24.7
24.6
24.0
130.5
20.3
15
26.6
29.5
25.7
25.8
24.7
18.7
16
123.3
123.8
124.3
123.8
16.9
25.8
17
132.2
130.5
129.9
130.5
123.7
18
18.3
17.1
24.2
23.9
130.5
19
17.9
16.9
16.9
16.9
24.6
20
16.9
OCH3-7
62.1
59.9
59.8
59.9
59.9
59.9
OCH3-11
59.4
a in CDCl3; b in CD3CD.
Compound 6 was obtained as a yellow solid, and its molecular formula was deduced to be C26H32O8 using HR-ESI-MS, which showed a molecular ion peak at m/z 472.2092 [M]+. It exhibited 1H and 13C NMR spectra similar to those of α-mangostin (1),18 suggesting 6 also comprised a 13,6-trihydroxy-7-methoxy-8-(3-methylbut-2-enyl)-9H-xanthen-9-one structure. Two additional methyls at δH 0.97 (3H, s, H-14) and 0.93 (3H, s, H-15)/δC 20.3 (C-14) and 18.7 (C-15), a methylene at δH 2.81 (2H, m, H-10)/δC 23.2 (C-10), an oxygenated methylene at δH 3.41 (2H, m, H-11)/δC 85.9 (C-11), and an oxygenated methane at δH 3.52 (2H, m, H-13)/δC 68.8 (C-13) indicated the presence of a 4-hydroxy-2-methoxy-3,3-dimethylbutyl group (Tables 1 and 2). This was supported by the 1H-13C HMBC correlations of H11/C-10, C-12, C-13, C-14, C-15; OCH3/C-11; and H-13/C-11, C-12, C-14, C-15 (Figure 3). Furthermore, the 4-hydroxy-2-methoxy-3,3-dimethylbutyl group was assigned to C-2 based on the 1H-13C HMBC correlations of H-10/C-1, C-2, C-3 and H-11/C-2. All 1H and 13C NMR signals of 6 were unambiguously confirmed by 1H-1H COSY, 1H-13C HSQC, and 1H-13C HMBC NMR data. Thus, compound 6 was elucidated as 1,3,6-trihydroxy-2-(4-hydroxy-2-methoxy-3,3-dimethylbutyl)-7-methoxy-8-(3-methylbut-2-enyl)-9H-xanthen-9-one, which is a previously unknown compound.
Proposed Mechanisms for the Generation of Products upon the Radiolysis of α-Mangostin in Methanol
α-Mangostin has a xanthone (9H-xanthene) structure substituted by 1,3,6-hydroxy, 7-methoxy, 9-oxo, and 2,8-prenyl groups. The 1-, 3-, and 6-hydroxyl substituents of α-mangostin facilitate redox reactions during radiolysis. γ-Radiolysis of α-mangostin in methanol generated ·OCH3 and ·CH2OH from methanol and produced several α-mangostin derivatives with the modified prenyl group at C-2. Based on 1D and 2D NMR data, we presume that the ·OCH3 or ·CH2OH radicals react with the hydroxy group localized to C-1 or C-3 to produce a 1-oxy-mangostin radical (M1) and CH3OH (Figure 4A). The transformation mechanism of mangostin (1) to 2 was proposed to occur via the formation of M2, M3, and M4 owing to the mesomeric effect. The same mechanism has been observed during kaempferol radiolysis.19 Based on the report on cyclization during the radiolysis of rosmarinic acid,20 we presumed that the CH· at C-12 in the prenyl group would react with the 3-OH to form the ring compound M5. Then, M5 would be hydrogenated by the hydrogen generated from methanol radiolysis, producing 2 (Figure 4A).
Proposed mechanism for the formation of compounds 2–6 upon the γ-irradiation of α-mangostin (1).
The formation mechanisms of the intermediate products M2 and M3 during the production of 3 are identical to that of 2 (Figure 4B). Next, during the production of 3, the M3 radical reacts with CH3OH, generating the intermediate product M6. Other ·OCH3 radicals would react with the -OCH3 group of M6, yielding the M7 radical and CH3CH2OH, which is the same mechanism as that reported for the generation of the intermediate products of kaempferol.19 Then, the lone pair on M7 would be transferred to C-12 of the prenyl group to form the M8 radical, which cyclizes with 1-OH (M9),20 and the double bond would be hydrogenated to form 3.
However, the mechanism of the formation of the M4 radical during the formation of 4 would be the same as that during the formation of 2 (Figure 4C). Next, the CH· at C-12 in the prenyl group of the M4 radical would react with CH3OH to form the M10 radical, and the ·OCH3 radicals would react with the -OCH3 group of M10, yielding the M11 radical and CH3CH2OH.20 The lone pair on M11 would be transferred to the C-11 of the prenyl group to form the M12 radical, which cyclizes with 3-OH (M13).20 Subsequently, the double bond of furan would be hydrogenated to form 4.
The mechanism for the formation of 5 includes hydrolysis and hydroxyethylation, which have been observed during chrysin radiolysis (Figure 4D).21 The prenyl group in the M2 radical would be hydrolyzed and cleaved to produce the M14 radical. Two ·CH2OH radicals would react with M14 to form 5.
The mechanism of the formation of the M6 radical during the formation of 6 would be the same as that during the formation of 2 (Figure 4E). Subsequently, another ·OCH3 radical would react with M6, yielding the M15 radical with the lone pair transferred to C-12 of the prenyl group. Then, the ·CH2OH radical would be hydroxymethylated at C-12 to form 6, which was previously reported for the mechanism of rosmarinic acid radiolysis.20
Years of research on the effects of gamma radiation on natural products and single compounds have enhanced our understanding of their radiolysis mechanisms.2 In this regard, theoretical G-value calculations, used to calculate radiochemical yields, are recommended for molecular transformations using ionizing radiation.22 G-value indicates the number of radiolysis products generated and decomposed per 100 eV of absorbed energy, which helps address issues such as yield and reproducibility of radiolysis products.22 However, our study, which predicted the mechanisms for generating the radiolysis products of α-mangostin based on literature reporting the proposed mechanisms for deriving radiolysis products of flavonoids and phenolic acids,19–21 has several limitations. Therefore, further study is required to evaluate compounds with structures similar to α-mangostin under the same conditions, or to modify them according to different solvents and radiation sources.
Analysis of Neuraminidase Inhibition Activity and Kinetics
As the five radiolysis products 2–6 isolated from the γ-irradiated α-mangostin (1) in methanol contain substituents on the A-ring of α-mangostin, their enzyme activities and inhibition modes were evaluated and compared. The enzyme assay involved measuring the increase in the fluorescence of 4-methyl-umberillferyl-α-d-N-acetylneuraminic acid, which would be hydrolyzed by neuraminidase. We previously reported the neuraminidase inhibitory activities of α-mangostin (IC50 value = 12.9 μΜ) and 11 prenylated xanthones.6 However, no previous studies have evaluated the neuraminidase inhibitory activities of compounds 2–6.
Compounds 2–6 exhibited dose-dependent inhibitory activity against neuraminidase with IC50 values of 16.0, 24.46, 7.84, 14.88, and 44.66 μΜ, respectively (Table 3). As the concentration of the inhibitor increased, the residual enzyme activity decreased rapidly (Figure 5A). Compound 4 was 1.6 and 2 times more effective than the starting compound, α-mangostin (IC50 = 12.9 μΜ) and the positive control, quercetin (IC50 = 14.46 μΜ), respectively. Meanwhile, other compounds showed lower inhibitory activities than α-mangostin and quercetin. The inhibitory mechanism of compound 4 was subsequently studied. Figure 5B shows a plot of residual enzyme activity versus enzyme concentration at different concentrations of 4. As the concentration of 4 increased, the linear gradient representing enzyme activity rapidly decreased, and a group of straight lines with a y-axis intercept of 0 was created. This indicates that 4 is a reversible inhibitor, which temporarily limits enzyme activity by non-covalently biding to the enzyme. The enzyme inhibition modes were analyzed using Lineweaver–Burk plots of 1/velocity (1/V) versus 1/[S] and Dixon plots of 1/V max versus the concentration of compounds. Compound 4 exhibited intersecting lines at the non-zero points on the y-axis, where the common intercept on the x-axis decreased with increasing substrate concentration, suggesting the presence of competitive inhibition (Figure 5C). Other compounds also showed similar inhibitory behaviors with competitive inhibition (Figure S15). The inhibitor constant values (Ki) of 1–6, determined from the Dixon plot (Figures 5D and S16) and indicating the concentration required to produce half maximum inhibition, were 6.04, 7.46, 11.25, 3.55, 7.23, and 23.62 μΜ, respectively (Table 2). For all competitive inhibitors, plotting 1/V against the inhibitor concentration at each substrate concentration showed that the family of intersecting lines converged above the x-axis, with the [I] value at the point of intersection being –K. Thus, compounds 1–6 competitively and non-covalently bound to the substrate, thereby inhibiting the action of the enzyme. However, if the concentration of the substrate is increased, the enzyme activity can be restored.
(A) Effect of compounds on neuraminidase-catalyzed hydrolysis of 4-methylumbelliferyl-α-d-N-acetylneuraminic acid. (B) Neuraminidase hydrolytic activity in the presence of compound 4 [0 μM, ●; 5 μM, ○; 10 μM, ▾; 20 μM, △]. (C) Lineweaver–Burk plot for the inhibition of the hydrolysis activity of neuraminidase by compound 4. Enzyme assay conditions: 0.125 mM 4-methylumbelliferyl-α-d-N-acetylneuraminic acid and 50 mM sodium acetate buffer (pH 5.0) at 37 °C. (D) Dixon plot for the inhibition of the hydrolysis activity of neuraminidase by compound 4.
Neuraminidase Inhibitory Activities of Compounds 1-6.
Compound
Neuraminidase (Clostridium perfringens)
IC50a (μM)
Type of Inhibition (Kib, μM)
1
12.90 ± 0.8c
competitive (6.04 ± 1.2)
2
16.0 ± 1.8
competitive (7.46 ± 0.7)
3
24.46 ± 2.1
competitive (11.25 ± 0.9)
4
7.84 ± 0.7
competitive (3.55 ± 1.1)
5
14.88 ± 0.2
competitive (7.23 ± 0.6)
6
44.66 ± 2.4
competitive (23.62 ± 2.7)
Quercetin
14.46 ± 0.4
NTd
a IC50 values of compounds represent the concentration that caused 50% enzyme activity loss.
b Values of inhibition constant.
c IC50 value of compound 1 reported in a previous study.3
d NT = not tested.
Comparing the neuraminidase inhibition activities of the 12 prenylated xanthones reported in our previous study with those of compounds 1−6,6 we found that compound 4 showed more potent neuraminidase inhibition than mangostenone F owing to the addition of one hydroxyl group at the 2-(prop-1-en-2-yl)-2,3-dihydrofuran moiety. Compounds 2 and 3, which contain a prenyl group fused to a pyran or dihydropyran ring, similar to those in 9-hydroxycalabaxanthone and mangostanol, exhibited similar neuraminidase inhibitory activities. Similar to the kinetic results of compounds 1−6, the 12 reported prenylated xanthones were also reversible inhibitors and showed competitive inhibition. Thus, the prenylated xanthones produced in this study can be presented as potent competitive inhibitors with reversible kinetics and can be referenced as lead compounds for treating the influenza virus. However, further study is needed to explore its full range of potential uses for the development of natural influenza virus inhibitor.
Conclusion
This study demonstrated that the radiolysis of α-mangostin (1) in a methanol solution produced five structurally modified xanthones, including two new compounds 5 and 6. Their mechanism was proposed the mesomerism in the conjugated system, hydrogenation, cyclization, hydrolysis, hydroxymethylation, or hydroxyethylation for the formation of five derivatives, by the free radicals produced from methanolic radiolysis. Comparing to the original compound, α-mangostin (1), mangostanin (4) showed higher neuraminidase inhibitory activity with competitive inhibition, suggesting to be a potential neuraminidase inhibitor for treating influenza. Our study suggests a unique approach to the synthesis of the prenylated xanthone using ionizing radiation and to facilitate the discovery of compounds with improved biological activity.
Materials and Methods
General Procedure
NMR spectra were recorded on a Bruker AM500 instrument (1H NMR at 500 MHz, 13C NMR at 125 MHz; Bruker Billerica, MA, USA) and a JEOL ECX-500 instrument (1H NMR at 500 MHz, 13C NMR at 125 MHz; JEOL, Tokyo, Japan). HR-ESI-MS were obtained using a JEOL JMS-700 instrument. HPLC was performed using an Hitach L-2455 quaternary pump and an Hitach L-2455 series diode array detector (Hitach High Technologies Corporation, Japan), equipped with a TSK-GEL ODF-100 V column (4.6 × 150 mm, 5 μm; TOSOH, Tokyo, Japan). It was controlled by EZChrom Elite software (Agilent Technologies Co., Santa Clara, CA, USA). A CombiFlash® Companion flash chromatography system (medium pressure liquid chromatography (MPLC); Teledyne ISCO, Lincoln, NE, USA) and a SepBox 2D-250 instrument (Sepiatec GmbH, Berlin, Germany) were used for the separation and purification of compounds.
Chemicals and Reagents
α-Mangostin (1) was isolated from the ethanol extract of the pericarps of G. mangostana, and its structure was identified via spectroscopic analysis and comparison with reported spectroscopic data.3,16 HPLC analysis confirmed that the purity of α-mangostin (1) was >98%. Methanol-d4 and CDCl3 were purchased from Cambridge Isotope Laboratory Inc. (Andover, MA, USA). Neuraminidase (E.C. 3.2.1.18. from Clostridium perfringens), 4-methylumbelliferyl-α-d-N-acetylneuraminic acid sodium salt hydrate, and quercetin (a positive control) were purchased from Sigma-Aldrich (St. Louis, MO, USA). All other chemicals and solvents used in this study were of analytical grade.
γ-Irradiation Conditions
γ-Irradiation was performed at ambient temperature using a [60Co] γ-irradiator (150 TBq capacity; ACEL, Ontario, Canada) at the Advanced Radiation Technology Institute, Korea Atomic Energy Research Institute (Jeongup, Korea). Dosimetry was performed using 5-mm diameter alanine dosimeters (Bruker Instruments, Rheinstetten, Germany). The dosimeters were calibrated against an international standard set by the International Atomic Energy Agency (Vienna, Austria). The sample solution (3 g of α-mangostin in 3 L of MeOH) was divided into 300 vials, which were then irradiated with 200 kGy (absorbed dose). The dose rate on the sample was 10 kGy/h. The irradiated methanolic solution was immediately evaporated to remove the solvent and then lyophilized.
Isolation of Products from the γ-Irradiated α-Mangostin
The γ-irradiated α-mangostin (1) was subjected to reversed-phase column chromatography using methanol/water (3:7 to 1:0, v/v) as the gradient solvent system to yield seven fractions (F01–F07). Fraction F01 (120 mg) was subjected to MPLC with a C18 reverse phase cartridge column (20 g, 40-53 μm) using methanol/water (3:7 to 1:0, v/v; 10 mL/min) to afford four sub-fractions (F0101–F0104). Sub-fraction F0102 was purified by MPLC with a C18 reverse phase cartridge column (4 g, 40-53 μm) using an isocratic solvent system of methanol/water (7:3, v/v) to yield 4 (15 mg, 0.5% w/w). Fraction F02 (452 mg) was subjected to MPLC with a C18 reverse phase cartridge column (20 g, 40-53 μm) using an isocratic methanol/water solvent system (7:3, v/v) to yield seven sub-fractions (F0201–F0207). Sub-fraction F0207 was purified using the SepBox 2D-250 system with a C4 reverse phase cartridge column (40-53 μm) using isocratic methanol/water (7:3, v/v) to yield 3 (10 mg, 0.33% w/w). Fraction F04 (65 mg) was subjected to MPLC with a C18 reverse phase cartridge column (4 g, 40-53 μm) and eluted with methanol/water (7:3, v/v), affording five sub-fractions (F0401–F0405). Sub-fraction F0403 was subjected to MPLC with a RediSep normal-phase silica gel cartridge column (12 g, 35-60 μm) using a gradient solvent system of hexane/acetone (4:1 to 1:1, v/v) and further purified by a SepBox-2D-250 system with a C4 reverse phase cartridge column (40-53 μm) using isocratic methanol/water (7:3, v/v) to yield 5 (6 mg, 0.2% w/w). Fraction F06 (120 mg) was fractionated using MPLC packed with a C18 reverse phase cartridge column (20 g, 40-53 μm) and isocratic methanol/water (7:3, v/v) into four sub-fractions (F0601−F0604). Sub-fractions F0601–F0603 were combined (70 mg), subjected to MPLC with a RediSep normal-phase silica gel cartridge column (4 g, 35-60 μm), and eluted with chloroform/ethyl acetate (4:1 to 1:1, v/v) to afford a fraction-enriched 2. Further purification using a SepBox 2D-250 instrument with a C4 reverse phase cartridge column (40-53 μm) and an isocratic elution of methanol/water (7:3, v/v) yielded 2 (7 mg, 0.23% w/w). Fraction F07 (150 mg) was subjected to MPLC packed with a C18 reverse phase cartridge column (20 g, 40-53 μm) and eluted with isocratic methanol/water (7:3, v/v) to yield four sub-fractions (F0701–F0704). Sub-fraction F0702 was further purified by SepBox 2D-250 with a C4 reverse phase cartridge column (40-53 μm) and isocratic methanol/water (8:2, v/v) to yield 6 (7 mg, 0.23% w/w).
Compound 1. Yellowish powder; 1H NMR 500 MHz and 13C NMR 120 MHz, see Table 1; ESI-MS m/z 410 [M]+; HR-ESI-MS m/z 410.1727; calcd for C24H26O6, 410.1729.
Compound 2. Yellowish powder; 1H NMR 500 MHz and 13C NMR 120 MHz, see Table 1; ESI-MS m/z 410 [M]+; HR-ESI-MS m/z 410.1727 [M]+; calcd for C24H26O6, 410.1729.
Compound 3. Yellowish powder; 1H NMR 500 MHz and 13C NMR 120 MHz, see Table 1; ESI-MS m/z 426 [M]+; HR-ESI-MS m/z 426.1680 [M]+; calcd for C24H26O7, 426.1679.
Compound 4. Yellowish powder; 1H NMR 500 MHz and 13C NMR 120 MHz, see Table 1; ESI-MS m/z 426 [M]+; HR-ESI-MS m/z 426.1678 [M]+; calcd for C24H26O7, 426.1679.
Compound 5. Yellowish powder; 1H NMR 500 MHz and 13C NMR 120 MHz, see Table 1; ESI-MS m/z 386 [M]+; HR-ESI-MS m/z 386.1368 [M]+; calcd for C21H22O7, 386.1366.
Compound 6. Yellowish powder; 1H NMR 500 MHz and 13C NMR 120 MHz, see Table 1; ESI-MS m/z 472 [M]+; HR-ESI-MS m/z 472.2092 [M]+; calcd for C26H32O8, 472.2097.
Assay for the Neuraminidase Inhibition Activity
The neuraminidase inhibition activity of compounds 2-6 was measured using a modified, previously described method.6,23 Briefly, 4-methylumbelliferyl-α-d-N-acetylneuraminic acid sodium salt hydrate (0.125 mM) in 50 mM of sodium acetate buffer (pH 5.0) was used as a substrate. Neuraminidase (0.1 U/mL) in acetate buffer was used as the enzyme. The isolated compounds were dissolved in methanol and diluted to the corresponding concentrations in acetate buffer. For the assay, neuraminidase was first added to 15 μL of the sample solution and 510 μL of buffer in a cuvette. Then, 60 μL of substrate was added to this mixture at 37 °C. 4-Methylumbelliferone was immediately quantified fluorometrically on a SpectraMax M2 multi-mode microplate reader (Molecular Devices, CA, USA). The excitation wavelength was 365 nm, and the emission wavelength was 450 nm. The IC50 value was defined as the concentration of the compound at which 50% neuraminidase inhibition was observed. Kinetic parameters were determined using the Lineweaver–Burk double reciprocal plot. The determination of enzyme activity (by fitting experimental data to the logistic curve obtained using Equation 1) involved recording initial velocity over a range of concentrations, and the data were analyzed using a nonlinear regression program (Sigma Plot, SPCC Inc., Chicago, IL). Quercetin was used as the positive control.
Statistical Analysis
The assay was performed in a single experiment, and the samples were analyzed in triplicate. The IC50 values for the compounds were calculated using Sigma Plot (SPCC Inc., Chicago, IL, USA) via dose–response analysis. The Lineweaver–Burk double reciprocal plot is indicated as determined by nonlinear regression using Sigma Plot (SPCC Inc., Chicago, IL, USA).
Supplemental Material
sj-doc-1-npx-10.1177_1934578X241283184 - Supplemental material for α-Mangostin Derivatives Produced by γ-Irradiation and Their Inhibitory Activities Against Influenza Virus Neuraminidase
Supplemental material, sj-doc-1-npx-10.1177_1934578X241283184 for α-Mangostin Derivatives Produced by γ-Irradiation and Their Inhibitory Activities Against Influenza Virus Neuraminidase by Ah-Reum Han, Hyun Sim Woo, Dae Wook Kim, Hyung-Won Ryu and Chang-Hyun Jin in Natural Product Communications
Footnotes
Acknowledgments
We acknowledge the support of the Radiation Fusion Research Division, Advanced Radiation Technology Institute, Korea Atomic Energy Research Institute, for providing the irradiation facility.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Research program of Korea Atomic Energy Research Institute, (grant number Project No. 523320-24).
ORCID iD
Chang-Hyun Jin
Supplemental Material
Supplemental material for this article is available online.
SongHYKimKIHanJM, et al.Ionizing radiation technology to improve the physicochemical and biological properties of natural compounds by molecular modification: a review. Radiat Phys Chem.2022;194:110013. doi: https://doi.org/10.1016/j.radphyschem.2022.110013
3.
GuoZ. The modification of natural products for medical use. Acta Pharmaceutica Sinica B. 2017;7:119-136. doi.org/10.1016/j.apsb.2016.06.003
4.
ChinYJungHChaiHKellerWJKinghornAD. Xanthones with quinone reductase-inducing activity from the fruits of Garcinia mangostana (mangosteen). Phytochemistry. 2008;69(3):754-758. doi: https://doi.org/10.1016/j.phytochem.2007.09.023
5.
HanARKimJALantvitDD, et al.Cytotoxic xanthone constituents of the stem bark of Garcinia mangostana (Mangosteen). J Nat Prod.2009;72(11):2028-2031. doi: https://doi.org/10.1021/np900517h
6.
RyuHWCurtis-LongMJJungS, et al.Xanthones with neuraminidase inhibitory activity from the seedcases of Garcinia mangostana. Bioorganic & Medical Chemistry. 2010;18(17):6258-6264. doi: https://doi.org/10.1016/j.bmc.2010.07.033
7.
AlamMRashidSFatimaK, et al.Biochemical features and therapeutic potential of α-Mangostin: mechanism of action, medicinal values, and health benefits. Biomed Pharmacother.2023;163:114710. doi: https://doi.org/10.1016/j.biopha.2023.114710
8.
NaumanMCJohnsonJJ. The purple mangosteen (Garcinia mangostana): defining the anticancer potential of selected xanthones. Pharmacol Res.2022;175:106032. doi: https://doi.org/10.1016/j.phrs.2021.106032
9.
JohnODMushunjeATSurugauNGuadRM. The metabolic and molecular mechanisms of α-mangostin in cardiometabolic disorders. Int J Mol Med.2022;50(3):120. doi: https://doi.org/10.3892/ijmm.2022.5176
10.
YangAHLiuCWuJKouXShenR. A review on α-mangostin as a potential multi-target-directed ligand for Alzheimer's disease. Eur J Pharmacol.2021;897:173950. doi:https://doi.org/10.1016/j.ejphar.2021.173950
11.
DoHTTChoJ. Mangosteen pericarp and its bioactive xanthones: potential therapeutic value in Alzheimer’s disease, Parkinson’s disease, and depression with pharmacokinetic and safety profiles. Int J Mol Sci.2020;21(17):6211. doi: https://doi.org/10.3390/ijms21176211
12.
SultanOSKantilalHKKhooSP, et al.The potential of α-mangostin from Garcinia mangostana as an effective antimicrobial agent–A systematic review and meta-analysis. Antibiotics. 2022;11(6):717. doi: https://doi.org/10.3390/antibiotics11060717
13.
TarasukMSongprakhonPChimmaPSratongnoPNa-BangchangKYenchitsomanusP. Alpha-mangostin inhibits both dengue virus production and cytokine/chemokine expression. Virus Res.2017;240:180-189. doi: https://doi.org/10.1016/j.virusres.2017.08.011
14.
PatilPAgrawalMAlmelkarS, et al.In vitro and in vivo studies reveal α-mangostin, a xanthonoid from Garcinia mangostana, as a promising natural antiviral compound against chikungunya virus. Virol J.2021;18(1):1-12. doi: https://doi.org/10.1186/s12985-021-01517-z
15.
SugiyantoZYohanBHadisaputroS, et al.Inhibitory effect of alpha-mangostin to dengue virus replication and cytokines expression in human peripheral blood mononuclear cells. Nat Prod Bioprospect.2019;9(5):345-349. doi: https://doi.org/10.1007/s13659-019-00218-z
16.
PandaKAlagarasuKPatilP, et al.In vitro antiviral activity of α-mangostin against dengue virus serotype-2 (DENV-2). Molecules. 2021;26(10):3016. doi: https://doi.org/10.3390/molecules26103016
17.
SiaGLBennettGJHarrisonLJSimK. Minor xanthones from the bark of cratoxylum cochinchinense. Phytochemistry. 1995;38(6):1521-1528. doi: https://doi.org/10.1016/0031-9422(94)00526-y
18.
SenAKSarkarKKMazumderPCBanerjiNUusuoriRHaseTA. The structures of garcinones a, b and c: three new xanthones from Garcinia mangostana. Phytochemistry. 1982;21(7):1747-1750. doi: https://doi.org/10.1016/s0031-9422(82)85052-8
19.
MarfakATrouillasPAllaisDChampavierYCallisteCADurouxJ. Radiolysis of kaempferol in water/methanol mixtures and evaluation of antioxidant activity of kaempferol and products formed. J Agric Food Chem.2003;51(5):1270-1277. doi: https://doi.org/10.1021/jf020836g
20.
JeongGHChoJJoC, et al.Gamma irradiation-assisted degradation of rosmarinic acid and evaluation of structures and anti-adipogenic properties. Food Chem.2018;258:181-188. doi: https://doi.org/10.1016/j.foodchem.2018.03.068
21.
SongHYKimHMMushtaqS, et al.Gamma-irradiated chrysin improves anticancer activity in HT-29 colon cancer cells through mitochondria-related pathway. J Med Food.2019;22(7):713-721. doi: https://doi.org/10.1089/jmf.2018.4320
22.
YamaguchiHUchihoriYYasudaNTakadaMKitamuraH. Estimation of yields of OH radicals in water irradiated by ionizing radiation. J Radiat Res.2005;46(3):333-341. doi: https://doi.org/10.1269/jrr.46.333
23.
PotierMMameliLBélisleM, et al.Fluorometric assay of neuraminidase with a sodium (4-methylumbelliferyl-alpha-D-N-acetylneuraminate) substrate. Anal Biochem.1979;94(2):287-296. doi.org/10.1016/0003-2697(79)90362-2
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.