Bioactive Pentacyclic Triterpenes from the Root Bark Extract of

Article Cite This: J. Nat. Prod. 2018, 81, 2169−2176

pubs.acs.org/jnp

Bioactive Pentacyclic Triterpenes from the Root Bark Extract of Myrianthus arboreus, a Species Used Traditionally to Treat Type‑2 Diabetes Pierre B. Kasangana,†,‡,§ Pierre S. Haddad,‡,§ Hoda M. Eid,‡,§,⊥ Abir Nachar,‡ and Tatjana Stevanovic*,†,§ †

Wood Chemistry Laboratory, Department of Wood Sciences, Université Laval, 1045 Québec G1 V 0A6, Canada Natural Health Products and Metabolic Diseases Laboratory, Department of Pharmacology and Physiology, Université de Montréal, Montréal H3C 3J7, Canada § Institute of Nutrition and Functional Foods, Université Laval, 2440 Boulevard Hochelaga, Québec City G1 V 0A6, Canada ⊥ Department of Pharmacy, Beni-Suef University, El-Shahid/Shehata Ahmed Hijazy St 62514, Beni-Suef, Egypt

Downloaded via UNIV OF TEXAS AT EL PASO on October 30, 2018 at 12:05:19 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

‡

S Supporting Information *

ABSTRACT: Four new Δ12 ursene-type pentacyclic triterpenes containing the trans-feruloyl moiety (1−4), along with ursolic acid (5), were isolated from a Myrianthus arboreus root bark ethanol extract, after bioassay-guided subfractionation of its hexane fraction. The structures of 1−4 were established on the basis of the results of standard spectroscopic analytical methods (IR, HRESIMS, GC-MS, 1D and 2D NMR). The compounds 3β-O-trans-feruloyl-2α,19α-dihydroxyurs-12-en-28-oic acid (1), 2αacetoxy-3β-O-trans-feruloyl-19α-hydroxyurs-12-en-28-oic acid (3), and 5 were determined to decrease the activity of hepatocellular glucose-6-phosphatase (G6Pase) and to activate glycogen synthase (GS). Their action on G6Pase activity implicated both Akt and AMPK activation. In addition, these compounds were determined to stimulate GS via the phosphorylation of glycogen synthase kinase-3. Compound 3 showed the most potent effect in modulating glucose homeostasis in liver cells. This is the first comprehensive report on novel phytochemical components of the root bark extract of M. arboreus based on the isolation of the principles responsible for its antidiabetic effects. extracts.15,16 We recently confirmed the antioxidant potential of M. arboreus root bark ethanol extract. Furthermore, a comprehensive approach with in vitro bioassays17 was used recently to report, for the first time, the antidiabetic effect of M. arboreus root bark. In this study, the ethanol (EtOH) extract as well as its hexane and ethyl acetate (EtOAc) fractions significantly regulated glucose homeostasis in liver cells (H4IIE and HepG2 hepatocytes). In contrast, only a mild effect on muscle glucose transport in differentiated C2C12 myocytes was observed.18 The ethanol extract and its fractions were determined to decrease hepatocyte glucose-6phosphatase (G6Pase) activity through mechanisms involving both the insulin-dependent Akt pathway and the insulinindependent pathway, implicating the phosphorylation of AMP-activated protein kinase (AMPK). On the other hand,

Myrianthus arboreus P. Beauv. (Cecropiaceae) is a species of the genus Myrianthus that grows in tropical regions. This indigenous tree of the secondary forests of West and Central Africa is found throughout the northwest of the Democratic Republic of Congo.1,2 It is used widely as a food as well as in folk medicine to treat symptoms associated with diabetes (notably, its root bark) and other diseases.1−4 As a member of the family Cecropiaceae, M. arboreus has been reported to contain oleanane- and ursane-type pentacyclic triterpenes including myriaboric acid, ursolic acid, euscaphic acid, tormentic acid, myrianthic acid, myrianthic acid, myrianthinic acid, arjulonic acid, arboreic acid, and their derivatives notably in the leaves, stem bark, and trunk wood.5−10 Likewise, peptide alkaloids such as myrianthines A, B, and C11 and phytosterols such as β-sitosterol and β-sitosterol-3-O-β-D-glucopyranoside12 were isolated from M. arboreus leaves. Previous investigations have shown the antibacterial,13 wound healing, anti-infective, antinociceptive,14 and antioxidant properties of M. arboreus © 2018 American Chemical Society and American Society of Pharmacognosy

Received: January 25, 2018 Published: October 18, 2018 2169

DOI: 10.1021/acs.jnatprod.8b00079 J. Nat. Prod. 2018, 81, 2169−2176

Journal of Natural Products

Article

Chart 1

M. arboreus preparations were found to exert a promising effect on glycogen synthase (GS) activity, by increasing glycogen synthase kinase-3 (GSK-3) phosphorylation.19 G6Pase and GS are the rate-limiting enzymes, involved in hepatic gluconeogenesis and glucose storage pathways, respectively. The liver is a crucial organ involved in systemic glucose homeostasis and, therefore, plays a major role in type-2 diabetes.20 In a continuation of these previous studies, the aim of the present investigation was to identify bioactive compounds present in M. arboreus root bark extract using hepatic glucose metabolism assays.17 Toward this aim, bioassay-guided fractionation of the highly active hexane fraction of the ethanolic extract was carried out in order to isolate and identify the bioactive molecules that could be responsible for the observed effect on G6Pase and GS activities as well as on main kinases modulating these two enzymes.

fractions (SFH1−SFH6) were thus obtained and tested by the same hepatic glucose metabolism assays. Fractions SFH3 and SFH4 significantly reduced G6Pase activity by 21% and 55%, respectively, and activated GS by 2.3- and 3.5-fold, respectively, as compared to DMSO (0.1%) (Figure S26, Supporting Information). Fractions SFH3 and SFH4 were analyzed initially by GC-MS and HPLC-HRTOFMS to identify some of the molecules that have been previously reported for the genus Myrianthus. These analytical methods yielded molecular ion and mass fragments corresponding to one pentacyclic triterpene (5) and two phytosterols, 6 and 7. Comparison of their mass spectra with those previously reported for compounds isolated from different parts of M. arboreus12 enabled the identification of ursolic acid (5), stigmasterol (6), and β-sitosterol (7). These structures were also confirmed by the GS-MS profiles of the authentic standards. However, compounds 1−4 presented molecular ion and mass fragments that could not be associated with any of the compounds previously identified in M. arboreus. Further purification was performed by semipreparative HPLC-DAD and resulted in the isolation of four main compounds of the hexane fraction, namely, 1 and 2 from subfraction SFH3 and 3 and 4 from SFH4, as described in the Experimental Section. The structures of compounds 1 to 4 were elucidated by 1D and 2D NMR experiments, and the comparison of their physical and spectroscopic data with those previously reported in the literature allowed their identities to be confirmed (Table 1). Compound 1 was isolated as a white powder and gave a [M + H]+ ion peak at m/z 665.3993 (calcd for C40H57O8+, m/z

■

RESULTS AND DISCUSSION The root bark powder of M. arboreus was extracted with 95% EtOH. The crude extract was dissolved in distilled water and successively fractionated with hexane and EtOAc. As observed previously, the effect of the EtOH extract in glucose homeostasis bioassays was associated primarily with its hexane fraction. This fraction was demonstrated to significantly inhibit the activity of hepatocellular G6Pase (36% inhibition) by increasing both Akt and AMPK phosphorylation (by 2− 2.5fold in comparison with vehicle control), while simultaneously stimulating GS activity (by 2.5-fold in comparison with vehicle control) through GSK-3 phosphorylation.18 For this reason, the hexane fraction was selected and submitted to bioassayguided fractionation using flash chromatography. Six sub2170



Article

Table 1. 1H NMR (700 MHz) and 13C NMR (140 MHz) Spectroscopic Data for Compounds 1−4 in DMSO-d6 1 position

δC

1ax 1eq 2 3 4 5 6ax 6eq 7ax 7eq 8 9 10 11ax 11eq 12 13 14 15ax 15eq 16ax 16eq 17 18 19 20 21ax 21eq 22ax 22eq 23 24 25 26 27 28ax 29 30 CH3OH-19 1′ 2′ 3′ 4′ 5′ 6′ 7′ 8′ 9′ OCH3-3′ HOC-4′ 1″ 2″

48.1 65.3 84.1 39.9 55.0 18.5 32.9 39.6 47.1 37.8 23.7 127.0 139.2 41.6 28.5 25.6 47.3 53.7 72.1 41.9 27.1 37.7 28.9 18.3 16.9 17.1 24.4 179.4 16.7 26.9 126.1 111.5 148.4 149.7 116.0 123.5 144.9 115.8 167.2 56.2

2

δH (J in Hz) 1.87, 0.96, 3.69, 4.53,

overlap overlap overlap d (9.8)

0.95, 1.48, 1.38, 1.51, 1.24,

brt (12.0) overlap overlap overlap overlap

1.71, overlap 1.93, overlap 0.90, overlap 5.19, brt (3.3)

1.70, 0.91, 2.50, 1.39,

overlap overlap td (12.1, 5) overlap

2.39, s 1.26, 1.24, 1.10, 1.61, 1.54, 0.82, 0.88, 0.97, 0.72, 1.32,

overlap overlap overlap overlap overlap s s s s s

0.85, d (6.3) 1.10, s

7.31, d (0.7)

6.80, 7.09, 7.54, 6.47,

d (7.7) dd (7.7, 0.7) d (15.4) d (16.1)

3.88, s

δC 48.0 65.3 84.1 39.9 55.0 18.5 32.9 39.6 47.1 37.8 23.7 127.1 139.2 41.6 28.5 25.6 47.3 53.6 72.1 41.8 28.0 37.7 28.9 18.3 16.7 17.1 24.4 179.5 16.7 26.9 55.1 126.1 111.4 148.4 149.7 115.9 123.5 144.8 115.8 167.2 56.2

3

δH (J in Hz) 1.87, 0.87, 3.69, 4.53,

overlap overlap overlap d (9.8)

0.95, 1.48, 1.38, 1.51, 1.24,


δC 43.9 69.3 80.2 39.6 54.1 18.3 32.2 39.6 46.9 38.1 23.3


1.70, 0.91, 2.50, 1.39,

overlap overlap td (11.7, 4.9) overlap

25.6 47.3 53.7 72.3 41.8 39.1

2.39, s 1.26, 1.21, 1.20, 1.60, 1.54, 0.82, 0.88, 0.97, 0.72, 1.32,

126.9 139.0 41.6 28.3


37.7 28.5 18.2 16.6 17.0 24.4 179.4

0.85, d (6.3) 1.10, s 3.77, s

26.8

7.31, d (0.7)

6.78., d (7.7) 7.09, dd (7.7, 0.7) 7.54, d (15.4) 6.47, d (16.1) 3.88, s

δH (J in Hz) 1.92, 1.14, 5.09, 4.80,

overlap overlap ddd (14.4, 10.4, 4.5) d (10.5)

1.07, 1.51, 1.14, 1.52, 1.26,



1.69, 0.93, 2.50, 1.42,

overlap overlap td (12.2, 4.8) overlap

2.38, s 1.25, overlap 1.43, td (13.5, 4) 1.20, overlap 1.61, overlap 1.5, overlap 0.86, s 0.94, s 1.03, s 0.72, s 1.33, s

1.08, s

126.1 111.5 148.4 149.6 115.9 123.8 145.6 114.5 166.9 56.1

3.82, s

170.8 21.2

1.85, s

7.33, s

6.78, 7.10, 7.53, 6.46,

d d d d

(7.7) (7.7) (16.1) (16.1)

4 δC 43.4 69.8 80.0 39.6 54.0 18.4 32.8 39.7 46.7 37.9 23.6 126.9 139.2 41.7 28.5 25.4 47.4 53.6 72.0 41.8 38.7 37.7 28.6 18.2 16.8 16.9 24.2 179.4 26.8 55.0 126.0 111.5 148.4 149.6 115.9 123.8 145.6 114.4 166.8 56.0 191.5 170.1 21.2

δH (J in Hz) 1.92, 1.14, 5.10, 4.80,

overlap overlap ddd (14.5, 10.4, 4.5) d (10.5)

1.09, 1.51, 1.14, 1.56, 1.25,



1.69, overlap 0.93 2.49, td (11.8, 5.1) 1.39, overlap 2.38, s 1.25, 1.25, 1.20, 1.60, 1.50, 0.87, 0.94, 1.02, 0.72, 1.33,


1.07, s 3.79, s 7.33, s

6.79, 7.10, 7.53, 6.46,

d d d d

(7.7) (7.7) (16.1) (16.1)

3.82, s 9.71, s 1.82, s

with tormentic acid previously isolated from M. arboreus.6 The 1 H and 13C NMR data revealed the presence of seven methyl groups including six tertiary methyls and one secondary methyl [(δH/δC 0.82/28.9, 0.88/18.3, 0.97/16.9, 0.72/17.1, 1.32/24.4, 0.85 (×2)/16.7] and one vinylic proton at δH 5.19 that appeared as a triplet (J = 3.3 Hz), compatible with the

665.4047), in the positive-ion HRESIMS, appropriate for a molecular formula of C40H56O8. The IR spectrum showed absorption bands at 3392 cm−1 (OH), 1696 cm−1 (CO), and 1631 cm−1 as well as 1515 cm−1 (aromatic CC), respectively. The analysis of the 1D NMR data for compound 1 (Table 1) indicated that it shares close structural similarities 2171



Article

Figure 1. Selected COSY and HMBC correlations for 1 and 3.

Figure 2. Effect of compounds isolated from the bioactive subfractions on G6Pase (A) and GS (B) activities. Data obtained represent the change in G6Pase (A) and GS (B) activity observed after 18 h of treatment of hepatic cells (H4IIE and HepG2) with the maximal nontoxic concentration of M. arboreus extract, the hexane fraction, and the five compounds (1−5) (Table S1, Supporting Information). Data are reported relative to DMSO (0.1%) used as vehicle control. Insulin (100 nM) and AICAR (2 mM) were used as a positive control. *p < 0.05, **p < 0.01, and ***p < 0.001 denote statistically significant from vehicle control (0.1% DMSO).

55.1) to δC 72.1 (C-19) (Table 1). Consequently, the structure of compound 2 was defined as 3β-O-trans-feruloyl-2α-hydroxy19α-methoxyurs-12-en-28-oic acid. Compound 3 was isolated as a white powder and gave a molecular formula of C42H58O9 as determined from the HRESIMS data (calcd for C42H60O10, m/z 724.4385, [M + H2O]+). The 1D NMR spectra of compound 3 displayed resonances for seven singlet methyl groups (δH/δC 0.72/17.0, 0.86/28.5, 1.33/24.4, 0.94/18.16, 1.03/16.6, 1.85/21.3), an olefinic proton [δH 5.16 (1H, brd, J = 3.5 Hz)], and an acetoxy group (δH 1.85, δC 21.2 and δC 170.8). These 1H and 13C NMR data, together with the molecular formula, suggested the presence of a pentacyclic triterpenic acid containing an acetoxy group. The connectivity of the fragments and functionalities were determined by the HMBC correlation. The long-range correlations of a methyl group signal (δH 1.85) with those of the carbonyl functionality (δC 170.8) of the acetoxy group and with C-2 (δC 69.3) helped locate the acetoxy group at C-2. The latter was also confirmed by the H-2 signal that appeared at δH 5.09 as a ddd (J = 14.4, 10.4, 4.5 Hz) and suggested H-2 to be β-axial. Accordingly, the acetoxy group at this position was α-equatorial in accordance with spectroscopic data of serratolide (2α-acetoxy-3β-hydroxyurs-12-en-19,28-olide) isolated from Myrianthus serratus.22 This was confirmed by the presence of a signal at δC 4.80, which appeared as a doublet (J = 10.5 Hz) attributed to H-3 at an α-axial position geminal to an additional trans-feruloyl moiety. The 1H NMR signals at δH 7.53 (1H, d, J = 15.4 Hz), 7.33 (1H, d, J = 0.7 Hz), 7.10 (1H, dd, J = 7.7, 0.7 Hz), 6.78 (1H, d, J = 7.7 Hz), 6.46 (1H, d, J = 16.1 Hz), and 3.82 (3H, s); δC 126.1, 111.5, 148.4, 149.6, 115.9, 123.8, 145.6, 114.5, 166.9, and 56.1 were consistent with a trans-feruloyl group. This relative configuration was

presence of an allylic methylene group at C-11 (δC 23.7), together with 30 carbon signals indicative of a triterpenoid skeleton. The presence of the secondary methyl group suggests that compound 1 belongs to the Δ12 ursene-type pentacyclic triterpenes, and the hydroxy group located at quaternary carbon C-19 (δC 72.1) confirmed the presence of a tormentic acid skeleton pattern.6 In addition, compound 1 was found to possess an additional trans-feruloyl moiety, which was recognized by the NMR chemical shifts: δH 7.31 (1H, d, J = 0.7 Hz), 6.80 (1H, d, J = 7.7 Hz), 7.09 (1H, dd, J = 7.7, 0.7 Hz), 7.54 (1H, d, J = 15.4 Hz), 6.47 (1H, d, J = 16.1 Hz), and 3.88 (3H, s); δC 126.1, 111.5, 148.4,149.7, 116.0, 123.5, 144.9, 115.8, 167.2, and 56.2. HMBC correlations of H-3 (δH 4.53) and H-7′ (δH 7.54) with C-9′ (δC 167.2) supported the feruloyl moiety being attached to the tormentic unit at C-3 (δC 84.1) (Figure 1). The trans-configuration was determined by the presence of vinylic protons at δH/δC 6.47/115.8 and 7.54/ 144.9, with a large coupling constant (J = 15.4 Hz) and also by an interaction between H-9′ (δH 6.4) and methoxy group 3′ (δH 3.88) revealed by a NOESY correlation. The overall 1D and 2D NMR data indicated that the structure of compound 1 is a geometric isomer of 3β-O-cis-feruloyl-2α,19α-dihydroxyurs-12-en-28-oic acid reported by Kim et al.21 Based on this evidence, the structure of compound 1 was seen as new and was consequently established as 3β-O-trans-feruloyl-2α,19αdihydroxyurs-12-en-28-oic acid. Compound 2 was determined to have a molecular formula of C41H58O8, on the basis of its HRESIMS data (calcd for C41H59O8Na, m/z 702.5302, [M + Na]+). The structure of compound 2 was found to be similar to that of compound 1, with the difference being that the methoxy group is connected to C-19 based on an HMBC correlation from δH 3.77 (δC 2172



Article

Figure 3. Effect of bioactive compounds isolated from the bioactive fractions on kinases modulating G6Pase and GS activities. (A) Akt phosphorylation. Phosphorylated (p-Akt) and total Akt were measured by Western immunoblot as described in the Experimental Section (Table S1, Supporting Information). (B) AMPK phosphorylation. Phosphorylated (p-AMPK) and total AMPK were measured by Western immunoblot as described in the Experimental Section. (C) GSK phosphorylation. Phosphorylated (p-GSK) and total GSK were measured by Western immunoblot as described in the Experimental Section. H4IIE and HepG2 hepatocytes were treated with the maximal nontoxic concentration of the three compounds (1, 3, 5), 100 nM insulin, and 2 mM AICAR. *p < 0.05, **p < 0.01, and ***p < 0.001 denote statistically significant from vehicle control (0.1% DMSO).

established on the basis of the 1H NMR coupling constants (1H, d, J = 16.1 Hz) and 1H−1H NOESY correlations. The trans-feruloyl group was linked to C-3, as confirmed by the HMBC correlations from the vinylic protons at δH 7.53 and δH 6.46 as well as from the proton signal at δH 4.80 (δC 80.2) to the carbon signal at δC 166.9 (C-9′) (Figure 1). These spectroscopic observations led to the assignment of structure 3 as 2α-acetoxy-3β-O-trans-feruloyl-19α-hydroxyurs-12-en-28oic acid. Compound 4 was obtained as a white powder, and the HRESIMS gave a molecular ion at m/z 732.9421 [M + H]+ (calcd for C44H59O9, 732.9418), corresponding to C43H60O9. The 1H and 13C NMR data of compound 4 were found to be closely comparable to those of compound 3, except for the presence of a formyl group resonance [δH 9.71 (1H, s)/δC 191.5] that replaced the phenolic hydroxy group at the additional trans-feruloyl moiety. On the basis of the HMBC spectrum, the long-range correlations of the formyl group proton with C-3′ (δC 148.4), C-4′ (δC 149.6), and C-5′ (δC 115.9) indicated it to be attached to C-4′ (Table 1, Figure 1). Therefore, the structure of 4 was established as 2α-acetoxy-3β-

O-trans-(3′-methoxy-4′-formyl)cinnamoyl-19α-methoxyurs12-en-28-oic acid. In order to evaluate the pharmacological effect of the isolated and newly characterized compounds (1−4), these were tested along with ursolic acid (5) at their maximal nontoxic concentration to investigate their effects on hepatic glucose metabolism in liver cells (H4IIE and HepG2 cells) (Table S1, Supporting Information). However, we did not evaluate the biological activity of 6 and 7, because these were found as minor components of the hexane fraction and were not soluble in DMSO, used as our universal solvent and negative control. Of the studied compounds, three (compounds 1, 3, 5) induced a significant reduction of G6Pase activity (ranging between 18% and 32%) in comparison to the 0.1% DMSO control (p ≤ 0.05; Figure 2A). Likewise, the same compounds (1, 3, 5) stimulated GS activity in HepG2 cells from 1.5- to 3.5-fold as compared to 0.1% DMSO (p ≤ 0.05; Figure 2B). In both bioassays, compounds 1, 3, and 5 are likely candidates as active principles responsible for the ability of M. arboreus to regulate glucose homeostasis in the liver cells. It is noteworthy that compound 3 (IC50 = 4.8 μM; EC50 = 4.1 μM, for G6Pase inhibition and GS stimulation, respectively) 2173



Article

which involves perturbations of liver Akt, AMPK, and GSK-3 phosphorylation.20 In this context, 2α-acetoxy-3β-O-transferuloyl-19α-hydroxyurs-12-en-28-oic acid (3) possessed the most potent biological activity in the context of liver cell glucose homeostasis and could serve to standardize M. arboreus root bark extracts to be used as an efficient agent against type-2 diabetes. Together with compounds 1 and 5, compound 3 could also be used as a marker for the authentication of natural products based on M. arboreus root bark extracts in the Democratic Republic of the Congo.

showed the most potent effect in all bioassays; these effects were close to those of the original EtOH extract (Figure S25, Supporting Information). In contrast, compound 1 was more active in reducing G6Pase activity, while ursolic acid (5) elicited a moderate effect in both bioassays. Several studies have reported that ursolic acid had exerted many potentially beneficial effects including anti-inflammatory in vivo and in macrophages in vitro, as well as a capacity to regulate lipid metabolism and allergic reactions.23,24 It is noteworthy that, to the best of our knowledge, none of these isolated molecules have been evaluated previously for effects on glucose homeostasis in the liver, and our study therefore brings forth their novel biological effects in this context. Furthermore, an effort was made to understand the molecular mechanisms of action underlying the effect of the active molecules on G6Pase and GS activities in hepatic cells. In terms of G6Pase activity, we determined the activity of the two key signaling pathways: the insulin-dependent Akt pathway and the AMPK pathway. Insulin served as a positive control for the first pathway, while 5-aminoimidazole-4carboxamide-1-β-D-ribofuranoside (AICAR) was applied as an activator of AMPK signaling. The plant ethanol extract, its hexane fraction, and three of the compounds (1, 3, 5) increased the ratio of p-Akt/total Akt by 2.2- to 4.8-fold in comparison with DMSO (Figure 3A). Compound 1 induced Akt phosphorylation more effectively than the other two active compounds (3, 5). However, no significant relationship was found between Akt activation and the inhibition of G6Pase (data not presented). Aside from the insulin receptor pathway, the insulin-independent AMPK pathway also helps to control gluconeogenesis and its key enzymes.25,26 In fact, the antidiabetic action of metformin, the most commonly prescribed oral antidiabetic agent, as well as that of many natural products is mediated through AMPK phosphorylation.27−29 The results of the present investigation show that all tested active compounds increased AMPK phosphorylation from 2.2- to 3.5-fold, compared to DMSO (p < 0.05; Figure 3B). The extent of AMPK phosphorylation instigated by compound 3 approached those of the EtOH extract and its hexane fraction of the plant. In contrast to Akt phosphorylation, a robust correlation was observed between the insulinindependent AMPK pathway and the reduction of G6Pase activity induced by all plant preparations and isolated compounds (R2 = 0.87; p < 0.05; Figure S25, Supporting Information). Finally, glycogen synthase kinase-3 was selected as a main pathway implicated in the activation of GS activity.19−30 Insulin and activated AMPK trigger the phosphorylation and inactivation of GSK-3, which leads to the simulation of glycogen synthase.19−30 The EtOH extract and the hexane fraction of M. arboreus, as well as compounds 3 and 5, increased the phosphorylation of GSK-3 similarly to AICAR (2-fold, used as positive control), whereas compound 1 was less potent, with a 1.8-fold rise in p-GSK-3/total GSK-3 ratio compared to DMSO (Figure 3C). Similar to what was observed with the activity of G6Pase, a significant correlation was found between the effect of M. arboreus extract samples on GS stimulation and their impact on GSK-3 phosphorylation (R2 = 0.67, p < 0.01, Figure S25, Supporting Information). Altogether, these results support the antidiabetic action of M. arboreus root bark extract as used in Congolese folk medicine.1−4 Indeed, exaggerated hepatic glucose production contributes significantly to hyperglycemia of type-2 diabetes,

■

EXPERIMENTAL SECTION

General Experimental Procedures. Optical rotations were determined using a P-1020 polarimeter (JASCO, Easton, MD, USA) with MeOH as solvent. IR spectra were recorded on a Bruker IFS-66/S Fourier-transform IR spectrometer (Bruker, Karlsruhe, Germany). Fractionation was performed using a Superflash 3100 preparative flash chromatographic system (Interchim, Montlucon Cedex, France), equipped with a quaternary pump, a PDA detector, and a fraction collector, at a flow rate of 20 mL min−1. HPLC analysis was carried out using a Zorbax SB-C18 column (250 mm × 4.6 mm, 5 μm), with an Agilent 1100 Series HPLC system (Agilent Technologies, Santa Clara, CA, USA) equipped with a quaternary pump, a diode-array detector (DAD), an autosampler, and a column compartment, at a flow rate of 12 mL min−1. Semipreparative HPLC was conducted on an Agilent 1260 Infinity apparatus (Agilent Technologies, Santa Clara, CA, USA) equipped with an autosampler, a quaternary high-pressure mixing pump, and a diode array detector, at a flow rate of 20 mL min−1 using a Zorbax SB-C18 column (250 mm × 21.2 mm, 7 μm). The 1D NMR and 2D NMR spectra were recorded in DMSO-d6 on a Bruker AVANCE III 700 NMR spectrometer (Bruker, Karlsruhe, Germany). HPLC-HRTOFMS was performed on an Agilent 6210 LC time-of-flight mass spectrometer with an ESI interface (Agilent Technologies, Palo Alto, CA, USA). The HRMS parameters were as follows: nitrogen flow rate, 6.0 L min−1; temperature, 325 °C; nebulizer, 30 psi; capillary voltage, 4000 V; fragmentor, 175 V; skimmer voltage, 65 V; and octopole radio frequency, 250 V. Detection was obtained in both negative and positive ion mode with an m/z range of 100−1000 and a scan time of 1 s. The mass spectrometer was used in the negative and positive modes using the Agilent ESI-L low concentration tuning mix (Agilent Technologies, Palo Alto, CA, USA). The gas chromatography−mass spectrometry (GC-MS) analyses were performed on a Varian system combining a CP-3800 GC and a Saturn 2200 MS/MS (Varian Inc., Walnut Creek, CA, USA) equipped with a Varian Factor Four capillary column (VF-5 ms 30 m × 0.25 mm) under the following conditions: carrier gas (helium), 5.0 bar; flow, 1 mL mim−1; split ratio set to 10; temperature gradient, 100−320 °C; rate, 5 °C/min; ion trap temperature, 250 °C; transfer line temperature, 300 °C; electronimpact ionization, 70 eV; scan range, m/z 50−650. Compounds were identified by comparing their mass spectra to those of existing databases (NIST 02, Adams and Essentia) and by comparing their retention times and mass spectra to those of the commercially available compounds: ursolic acid (90%), stigmasterol (95%) purchased from Sigma-Aldrich (St. Louis, MO, USA)m and βsitosterol (95%) from Herbstandard Inc. (Champaign, IL, USA). Plant Material. The root bark of Myrianthus arboreus was collected in south of Kinshasa, Democratic Republic of the Congo, in April 2017 and identified by Jean-Pierre Habbari, a botanist at University of Kinshasa. A voucher specimen is stored at the herbarium of this University (MM0181). The M. arboreus root bark was ground, and the thus obtained powder was carefully sealed and transported to Laval University (Quebec City, Canada) prior to laboratory work, as described previously.18 Extraction and Isolation. The pulverized and air-dried root bark (50 g) of M. arboreus was macerated three times with 95% ethanol for 24 h at room temperature, and the combined extracts were filtered.18 The EtOH filtrate was concentrated under a rotary evaporator, 2174



Article

cell lines) were grown in 12-well plates until confluence and treated for 18 h with the samples in various concentrations and with 0.1% DMSO (as the vehicle control). Cell culture media were collected separately for each condition (lactate dehydrogenase (LDH) released by damaged cells), and then cells were lysed by adding culture medium with 1% Triton X-100 (to completely release remaining intracellular LDH). The cytotoxicity detection kit (Roche, Mannheim, Germany) was used to determine LDH activity in the medium (released LDH) and Triton X-100 lysates (cellular LDH). Absorbance was measured (Wallac Victor2, PerkinElmer, Waltham, MA, USA) at a 590 nm wavelength. The ratio of released LDH to total LDH (total LDH = released LDH + cellular LDH) was determined for each, and the results were normalized to values obtained from cells treated with 0.1% DMSO. Glucose-6-phosphatase Assay. G6Pase was assessed in the H4IIE rat hepatoma cell line, which expresses an optimal activity of this enzyme.17 The inhibitory effects of plant preparations and compounds on this enzyme were determined following the procedure developed previously.18 Then common 90% confluent cells in 12-well plates were treated overnight with a negative control (0.1% DMSO vehicle), positive control (insulin, 100 nM), plant extracts, and isolated compounds at their respective maximal nontoxic concentrations. Cells were rinsed and lysed in phosphate buffer containing 0.05% Triton. Cell lysates were incubated in a glucose-6-phosphatecontaining buffer for 40 min at 37 °C. Supernant glucose generated in this reaction was determined using the Autokit glucose test kit (Wako Diagnostics, Richmond, VA, USA). Glucose levels were calculated using a standard curve obtained in parallel. Protein content was measured using the BCA method, and the activity of G6Pase was expressed relative to protein content. Results are presented as percentage changes from vehicle control activity (0.1% DMSO). Glycogen Synthase Assay. GS activity was evaluated by applying a modification of a method reported by Thomas et al. in confluent HepG2 human hepatoma cells, known to exhibit an optimal expression of GS.32,33 Treatments were applied for 18 h. Cells were stimulated for 15 min with 100 nM insulin used as a positive control. For the assay, 30 μL of supernatant from each condition was added to 100 μL of GS buffer (25 mM Tris, 5 mM UDP-glucose, 0.12 μ Ci/mL U−14C UDP-glucose, 1% glycogen, 3 mM EDTA, pH 7.9) in the presence (total form GS) or absence (active form GS) of 5 mM glucose-6-phosphate.34 After incubation, 90 μL of each was spotted onto filter paper squares (Whatman 31ET), which were rapidly submerged in ice-cold 66% ethanol and rinsed twice with 66% ethanol to remove unincorporated substrate from precipitated glycogen. Dried filters were then placed in scintillation vials. The resulted radioactivity for 14C was measured using a β-counter (LKB Wallac 1219; PerkinElmer, Woodbridge, ON, Canada). Activity of GS was reported as the activity ratio of the active form to total GS. Western Blot Analysis. For Western blot analysis, confluent cells (H4IIE and HepG2) were treated with samples or vehicle (DMSO) for 18 h, as described previously.33−35 Cells were resuspended in RIPA lysis buffer (pH 7.4) and centrifuged, and the supernatants collected. After quantification by BCA, 40 μg of protein from each solution was separated on gel by electrophoresis and transferred onto a nitrocellulose membrane (Millipore, Bedford, MA, USA). Subsequently, the membrane was blocked with 5% milk and incubated with the following primary antibodies: p-GSK-3 (Ser 9) (1:1000, 5% BSA, Millipore), GSK-3 (1:2000, 5% BSA, Millipore); phospho-Akt (Thr 308) (1:1000, 5% BSA, Cell Signaling Technology, Danvers, MA, USA), Akt (1:2000, 5% BSA, Cell Signaling Technology); phospho-AMPKα (Thr 172) (1:350, 5% BSA, Cell Signaling Technology), AMPK (1:1000, 5% BSA, Cell Signaling Technology). After rinsing in phosphate-buffered saline, the membranes were incubated with horseradish-peroxidase-conjugated secondary antibodies (anti-rabbit, Jackson, 1:20000), for 1 h, at room temperature. The bound antibodies were detected with HRPconjugated secondary antibodies, revealed by enhanced chemiluminescent reagent (Millipore, Billerica, MA, USA). Signals were measured using NIH ImageJ 1.45s software (National Institutes of

yielding a gummy brownish solid (1 g). The residue was dissolved in H2O and fractionated sequentially with hexane and EtOAc using a continuous liquid−liquid extractor.15 As reported previously, the activity of the EtOH extract using glucose homeostasis bioassays was concentrated in the hexane fraction.18 Therefore, the hexane fraction (450 mg) was separated by flash chromatography on a column (50 g) (Super-Fash SF25-80G, Si 50 PNAX1213-8) using a gradient solvent system of CHCl3−MeOH (100:0:0 to 1:2:0.1), for 23 min. This yielded five main subfractions: SFH1 (50 mg), SFH2 (90 mg), SFH3 (80 mg), SFH4 (190 mg), and SFH 5 (20 mg). These subfractions were tested by the same bioassays used previously (namely, G6Pase and GS assays). The most bioactive subfractions, SFH3 and SFH4, were then studied by analytical HPLC (Agilent Zorbax SB-C18 column 250 × 4.6 mm i.d., 5 μm; Agilent Technologies, Santa Clara, CA, USA). Their chromatograms exhibited four similar major peaks. SFH 3 (80 mg) was thus purified by semipreparative HPLC using 1% HCOOH in water (A) and CH3CN (75%, B) as mobile phase, 0−35 min, to gather the first main compounds 1 (32 mg, tR 21.5 min) and 2 (20 mg, tR 22.3 min). Compound 5 was isolated as a minor compound (1 mg, tR 7.1 min) from the same subfraction. Likewise, SFE 4 (190 mg) was purified employing semipreparative HPLC, with 1% HCOOH in water (A) and CH3CN (85%, B), 0−23 min, to furnish compounds 3 (45 mg, tR 8.5 min) and 4 (15 mg, tR 11.7 min). In addition, SFH3-4 (10 mg mL−1) was injected into the GC-MS through a goose-neck insert maintained at 280 °C, with splitless injection. The helium flow rate was kept at 1 mL min−1. The initial temperature was set at 100 °C, ramped up to 280 °C at rate of 25 °C min−1, then raised 5 °C min−1 to a final temperature of 325 °C, where it was held for 13.8 min. The compound 5 (tR 8.8 min) and phytosterols 6 (tR 14.2 min) and 7 (tR 15.3 min) were identified by comparing the retention times and mass spectra to the Adams and NIST02 databases and to those of the corresponding standards. 3β-O-trans-Feruloyl-2α,19α-dihydroxyurs-12-en-28-oic acid (1): white powder; [α]25 D +34.0 (c 0.1, MeOH); IR (KBr) νmax 3392, 2950, 1696, 1631, 1515, 1264 cm−1; 1H (700 MHz) and 13C (140 MHz) NMR data in DMSO-d6, see Table 1; HRESIMS m/z 665. 3993 [M + H]+ (calcd for C40H55O8, 665.4047). 3β-O-trans-Feruloyl-2α-hydroxy-19α-methoxyurs-12-en-28-oic acid (2): white powder; [α]25 D +18.0 (c 0.1, MeOH); IR (KBr) νmax 3392, 2938, 1696, 1631, 1515, 1264 cm−1; 1H (700 MHz) and 13C (140 MHz) NMR data in DMSO-d6, see Table 1; HRESIMS m/z 702.5301 [M + Na]+ (calcd for C41H59O8Na, 702.5302). 2α-Acetoxy-3β-O-trans-feruloyl-19α-hydroxyurs-12-en-28-oic acid (3): white powder; [α]25 D +20.0 (c 0.1, MeOH); IR (KBr) νmax 3369, 2 293, 1686, 1631, 1515, 1274 cm−1; 1H (700 MHz) and 13C (140 MHz) NMR data in DMSO-d6, see Table 1; HRESIMS m/z 724.4384 [M + H2O]+ (calcd for C42H60O10, 724.4385). 2α-Acetoxy-3β-O-trans-(3′-methoxy-4′-formyl)cinnamoyl-19αmethoxyurs-12-en-28-oic acid (4): white powder; [α]25 D +44.0 (c 0.1, MeOH); IR (KBr) νmax 3369, 2293, 1686, 1631, 1515, 1274 cm−1; 1H (700 MHz) and 13C (140 MHz) NMR data in DMSO-d6, see Table 1; HRESIMS m/z 732.9421 [M + H]+ (calcd for C44H60O9, 732.9418). Cell Culture. Hepatic cells (H4IIE and HepG2 cell lines) were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). Other reagents were purchased from Invitrogen Life Technologies (Burlington, ON, Canada), Wisent (St. Bruno, QC, Canada), and Sigma-Aldrich (Oakville, ON, Canada), unless otherwise specified as below. All cells were cultured routinely in a 5% CO2 incubator at 37 °C until 85−95% confluent and then treated with M. arboreus preparations for experiments described. H4IIE and HepG2 cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and in DMEM/F12 (50/50) medium (containing 10% FBS and 0.5% PS), respectively. Cytotoxicity Assays. Cytotoxicity assays were carried out to determine the maximal nontoxic concentrations of M. arboreus extracts and test compounds, based on the published method of Decker and Lohmann-Matthes.31 Hepatic cells (H4IIE and HepG2 2175



Article

(12) Ojinnaka, C. M.; Kenne, L. J. Nat. Prod. 1985, 6, 1002−1003. (13) Seukep, J. A.; Ngadjui, B.; Kuete, V. SpringerPlus 2015, 4, 1−6. (14) Olonode, E. T.; Aderibigbe, A. O.; Bakre, A. G. J. Ethnopharmacol. 2015, 171, 94−98. (15) Kasangana, P. B.; Haddad, P. S.; Stevanovic, T. Antioxidants 2015, 2, 410−426. (16) Awounfack, C. F.; Ateba, S. B.; Zingue, S.; Mouchili, O. R.; Njamen, D. J. Ethnopharmacol. 2016, 194, 169−178. (17) Haddad, P. S.; Musallam, L.; Martineau, L. C.; Harris, C.; Lavoie, L.; Arnason, J. T.; Foster, B.; Bennett, S.; Johns, T.; Cuerrier, A.; Coon Come, E.; Coon Come, R.; Diamond, J.; Etapp, L.; Etapp, C.; George, J.; Husky Swallow, C.; Husky Swallow, J.; Jolly, M.; Kawapit, A.; Mamianskum, E.; Petagumskum, J.; Petawabano, S.; Petawabano, L.; Weistche, A.; Badawi, A. Evid.-Based Complement. Alternat. Med. 2012, 2012, 893426. (18) Kasangana, P. B.; Nachar, A.; Eid, H. M.; Stevanovic, T.; Haddad, P. S. J. Ethnopharmacol. 2018, 211, 117−125. (19) Lee, J.; Kim, M. S. Diabetes Res. Clin. Pract. 2007, 77, S49−S57. (20) Roden, M.; Bernroider, E. J. Clin. Endocrinol. Metab. 2003, 3, 365−383. (21) Kim, C. S.; Subedi, L.; Oh, J.; Kim, S. Y.; Choi, S. U.; Lee, K. R. J. Nat. Prod. 2017, 80, 1134−1140. (22) Lontsi, D.; Ngounou, N. F.; Tapondjou, A. L.; Sondengam, B. L.; Bodo, B.; Martin. Phytochemistry 1998, 8, 2473−2476. (23) do Nascimento, P. G.; Lemos, T. L.; Bizerra, A. M.; Arriaga, A. M.; Ferreira, D. A.; Santiago, G. M.; Braz-Filho, R.; Costa, J. G. Molecules 2014, 19, 1317−1327. (24) Liu, J. J. Ethnopharmacol. 1995, 2, 57−68. (25) Bijland, S.; Mancini, S. J.; Salt, I. P. Clin. Sci. 2013, 8, 491−507. (26) Fogarty, S.; Hardie, D. G. Biochim. Biophys. Acta, Proteins Proteomics 2010, 3, 581−591. (27) Eid, M. H.; Haddad, S. P. Adv. Endocrinol. 2014, 2014, 1−11. (28) Cieniak, C.; Walshe-Roussel, B.; Liu, R.; Muhammad, A.; Saleem, A.; Haddad, P. S.; Cuerrier, A.; Foster, B. C.; Arnason, J. T. J. Pharm. Pharm. Sci. 2015, 18, 484−493. (29) Spoor, D. C.; Martineau, L. C.; Leduc, C.; BenhaddouAndaloussi, A.; Meddah, B.; Harris, C.; Burt, A.; Fraser, M. H.; Coonishish, J.; Joly, E.; Cuerrier, A.; Bennett, S. A.; Johns, T.; Prentki, M.; Arnason, J. T.; Haddad, P. S. Can. J. Physiol. Pharmacol. 2006, 84, 847−858. (30) Lawrence, J. C., Jr.; Roach, P. J. Diabetes 1997, 46, 541−547. (31) Decker, T.; Lohmann-Matthes, M.-L. J. Immunol. Methods 1988, 115, 61−69. (32) Eid, H. M.; Nachar, A.; Thong, F.; Sweeney, G.; Haddad, P. S. Pharmacogn. Mag. 2015, 11, 74−81. (33) Nachar, A.; Vallerand, D.; Musallam, L.; Lavoie, L.; Badawi, A.; Arnason, J.; Haddad, P. S. Evid.-Based Complement. Alternat. Med. 2013, 2013, 189819. (34) Danforth, W. H. J. Biol. Chem. 1965, 240, 588−593. (35) Shang, N.; Saleem, A.; Musallam, L.; Walshe-Roussel, B.; Badawi, A.; Cuerrier, A.; Arnason, J. T.; Haddad, P. S. PLoS One 2015, 10, e0135721.

Health, Bethesda, MD, USA). All experiments were performed on three independent cell passages. Statistical Analysis. The obtained results are presented as means ± SEM of three different experiments with triplicate determinations for each sample. Statistical calculations were performed by one-way analysis of variance (ANOVA) using Stat View software (SAS Institute Inc., Cary, NC, USA) and post hoc Bonferroni’s test. The IC50 or EC50 values were calculated using GraphPad Prism version 6 (GraphPad Software Inc., La Jolla, CA, USA). The level of significance was set at a p value of

Bioactive Pentacyclic Triterpenes from the Root Bark Extract of

Recommend Documents