Chemical Constituents of the Bark of Dipteryx alata Vogel, an Active Species against Bothrops jararacussu Venom

The effect of four sub-extracts prepared from the lyophilized hydroalcoholic bark of Dipteryx alata (Leguminosae-Papilionoideae) dissolved in a methanol-water (80:20) mixture through a liquid-liquid partition procedure has been investigated against the neuromuscular blockade of the venom of the snake Bothrops jararacussu. The active CH2Cl2 sub-extract has been extensively analyzed for its chemical constituents, resulting in the isolation of four lupane-type triterpenoids: lupeol (1), lupenone (2), 28-hydroxylup-20(29)-en-3-one (3), betulin (4), nine isoflavonoids: 8-O-methylretusin (5), 7-hydroxy-5,6,4’-trimethoxyisoflavone (6), afrormosin (8), 7-hydroxy-8,3’,4’-trimethoxyisoflavone (9), 7,3’-dihydroxy-8,4’-dimethoxyisoflavone (10), odoratin (11), 7,8,3’-trihydroxy-4’-methoxyisoflavone (13), 7,8,3’-trihydroxy-6,4’-dimethoxyisoflavone (15), dipteryxin (17), one chalcone: isoliquiritigenin (7), one aurone: sulfuretin (14) and three phenolic compounds: vanillic acid (12), vanillin (16), and protocatechuic acid (18). Their chemical structures were elucidated on the basis of spectroscopic analysis, including HRMS, 1D- and 2D-NMR techniques.


Introduction
Dipteryx alata Vogel, belongs to the Leguminosae-Papilionoideae family, and is a native species of the Brazilian Savanna, found principally in Minas Gerais, Goiás, Federal District and Mato Grosso. This species is known popularly as "baru" in Minas Gerais, "barujo" and "combaru" in Mato Grosso, and "cumaru" in other states. The plant has great economic potential because of its multiple uses. The fruit, rich in proteins, fibres and unsaturated fatty acids, is used as human and animal food, and also in cosmetic formulations [1][2][3]. Traditionally, the local inhabitants use the oil extracted from the almond to treat high fevers, and also for snakebites.
No extensive phytochemical investigations on Dipteryx alata were found, and the only previous report on the composition of this plant described the isolation of three lupane triterpenoids [4]. Previously, we have described the activity of D. alata methanolic extract obtained by Soxhlet extraction against the neuromuscular blockade of Bothrops jararacussu venom [5] from the pharmacological point of view. Exploiting the potential of the lyophilized D. alata hydroalcoholic extract, a liquid-liquid partition procedure based on varying polarity was carried out. This paper describes the results of a new investigation of the baru bark against the neuromuscular blockade produced by the jararacuçu snake (Bothrops jararacussu). Baru bark sub-extracts were obtained using apolar to polar types of solvents (hexane, dichloromethane, ethyl acetate and methanol); the efficacy of each was evaluated against muscle tissue damage and the neuromuscular blocking effects of Bothrops jararacussu venom.
The CH 2 Cl 2 sub-extract was one of the most active. For this reason, we have examined the constituents of this fraction, which has led to the isolation of four lupane-type triterpenoids 1-4, nine isoflavonoids 5, 6, 8-11, 13, 15, 17, one chalcone 7, one aurone 14 and three phenolic compounds 12, 16 and 18 ( Figure 1). All these compounds, with the exception of triterpenoids 1, 2 and 4, have been isolated from D. alata for the first time. The isoflavone 6 had been previously synthesized and isolated by acid hydrolysis of an isoflavone glycoside from other source, while it is being reported here as a natural aglycone. The isoflavone 13 had been isolated previously as an acetate derivative, and it is reported here as a natural product for the first time. Full spectroscopic data for the isolated isoflavonoids, including 13 C-NMR assignments, previously not described in literature, are reported herein.
Compound 6 gave a pseudomolecular ion peak at m/z 351.0847 [M + Na] + by HR-ESI-MS consistent with the molecular formula C 18 H 16 O 6 Na. The 1 H-NMR spectrum (Table 1) showed a singlet for H-2 at δ 7.80, characteristic of an isoflavone, another singlet at δ 6.78 corresponding to H-8 showed HMBC correlations with the signals of quaternary carbons at δ 153.8 (C-7), 154.6 (C-9) and 138.2 (C-6). Two doublets of double intensity at δ 6.96 and 7.47 confirmed the presence of a psubstituted ring B. Three O-methyl singlets at 3.84, 3.95 and 4.03 showed HMBC correlations with 13 C signals at 159.4 (C-4'), 151.7 (C-5) and 138.2 (C-6) respectively, indicating the location of the Omethyl groups. The 13 C-NMR (Table 2) data are in accordance with those reported for the aglycone obtained by acid hydrolysis of an isoflavonoid glycoside isolated from Baphia bancoensis [12], and also with a synthetic compound [11].   (Table 3) showed a singlet at 8.17 for H-2 typical of an isoflavone. The aromatic region of the spectrum displayed an A 2 B 2 -system (two doublets at δ 6.94 and 7.57, J = 8.7) characteristic for the A-ring in 7,8-dioxygenated isoflavones, and signals for three protons at δ 6.93, 6.96 and 7.03 of the disubstituted B-ring. From the above data and the 13 C-NMR signals (Table 4), this compound could be identified as 7,8,3'-trihydroxy-4'-methoxyisoflavone. To confirm this structure, compound 13 was acetylated. The 1 H-NMR spectrum of the acetyl derivative, 13a, showed three singlets at δ 2.29, 2.36 and 2.42 corresponding to the tri-O-acetyl derivative. Other 1 H-NMR data for 13a (Table 3) were in good agreement with those reported for an isoflavone isolated as acetate from Xanthocercis zambesiaca [17].

Anti-venom Assays
The protective activity of the four sub-extracts (hexane, dichloromethane, ethyl acetate and residual methanol) of the EtOH/H 2 O extract from Dipteryx alata, against the neuromuscular blockade of jararacuçu venom (Bothrops jararacussu, Bjssu venom), was confirmed by biological assays.

Mouse phrenic nerve-diaphragm (PND) preparation
The phrenic nerve-diaphragm muscle was obtained from mice previously anesthetized with halothane and sacrificed by exsanguination. The diaphragm was removed and mounted as described by Bülbring [23]. PND preparations were allowed to stabilize for at least 20 min before addition of one of the following solutions: Tyrode solution (control, n = 5); 40 µg/mL Bjssu (n = 6) venom, or by the mixture of venom plus 50 µg/mL of each extracts obtained by liquid partition [hexane, Hex (n = 4); dichloromethane, Dcm (n = 4); ethyl acetate, Eac (n = 5) or methanol, Met; n = 4] 30 min before addition to the neuromuscular preparation. The sub-extract concentration was chosen since did cause no effect on the basal response of preparation.
The Figure 2 shows the pharmacological profile of the in vitro neutralization assays after mixing 50 µg/mL of each extract (Hex, Dcm, Eac, and Met) with Bjssu venom (40 µg/mL) prior to test. Note that hexane (Hex) did not protect (5.2% ± 3.7) against irreversible neuromuscular blockade induced by crude venom (5% ± 5.6). A fair protection was seen with the ethyl acetate sub-extract (80% ± 6), whereas a practically total protection was seen with dichloromethane (98% ± 8) and methanol (95% ± 4.7) sub-extracts (* p < 0.05 compared to venom). As it can be seen, the Dcm and Met extracts showed no significant difference in relation with Tyrode control.  These results represent an attempt at identifying the active compounds responsible by the protective action of D. alata against Bothrops jararacussu venom, since preliminary data obtained from thin layer chromatography comparison pointed out to phenolic compounds present in the D. alata hydroalcoholic extract [5].
According to the screening of some chromatography fractions of the Dcm sub-extract, the F7 fraction displayed the most efficacious protection against the venom effect (results not shown). Taken these results together, compounds 12 (vanillic acid), 13 (3',7,8-trihydroxy-4'-methoxyisoflavone) and 14 (3',4',6-trihydroxyaurone), found in that fraction, could be the main responsible substances for the observed protection. Additionally, it can be stated that the triterpenoids are not responsible for the anti-Bothrops jararacussu venom activity, because they are also abundant in the inactive hexane sub-extract.
The mechanism by which these phytochemicals could neutralize the crude venom is yet unclear and further studies should be done. In fact, for investigating the mechanism of action or the pharmacological response, each isolated compound should be individually assayed against the venom, a condition not always possible due to the small isolated amounts. Besides, isolated compounds not always preserve the same pharmacological efficacy as that seen with total extract, as already related by elsewhere using the same experimental model [24,25].

General
IR spectra were obtained on a Nicolet Impact 410 spectrophotometer. 1 H-NMR (200 and 400 MHz), and 13 C-NMR (100 and 50 MHz) spectra were recorded on Bruker AC 200 (200 MHz) and Bruker DRX 400 (400 MHz) spectrometers, with TMS as an internal reference, δ given in ppm and J in Hz. 2D-NMR spectra were measured with Brüker DRX 400 spectrometer. HR ESI MS analysis, a VG-TS250 mass spectrometer (70 eV) was used. GC-MS analysis was performed on a Hewlett Packard 5890 series II, chromatograph apparatus equipped with Hewlett-Packard 5971 mass spectrometer operating in the EIMS-mode at 70 eV and Supelco SPB-1 column (12 m × 0.20 mm, with 0.33 µm film thickness) with helium (10 Psi), as carrier gas at a flow rate of 1 mL/min. The GC oven temperature was kept at 90 °C for 5 minutes and programmed to 250 °C with a gradient of 2 °C /min and maintained for 10 min at 250 °C, Wiley 275 as research library. Silica gel (230-400 mesh) for column chromatography and GF254 for TLC were obtained from Merck KGaA, 64271 Darmstadt, Germany. Sephadex LH-20 was obtained from Fluka, BioChemika Switzerland.

Statistical analysis
The results were expressed as the mean ± S.E.M. Student's t-test was used for statistical comparison of the data and the significance level was set at 5%.

Plant material
The bark of Dipteryx alata was collected in Pedro Afonso in the Tocantins Estate of Brazil, in August of 2007. The plant was identified by Dr. Roseli B. Torres from "Núcleo de Pesquisa e Desenvolvimento do Jardin Botânico", Institute of Agronomy of Campinas. A voucher specimen: IAC 50629 was deposited in the Institute of Agronomy of Campinas (Brazil).

Animals
Male Swiss white mice (2632 g) were supplied by Anilab: Animais de Laboratório (Paulínia, S.P., Brazil). The animals were housed at 25 ± 3 °C on a 12 h light/dark cycle and had access to food and water ad libitum. This project (protocol n o A012/CEP/2006) was approved by the institutional Committee for Ethics in Research of Vale do Paraiba University (UNIVAP), and the experiments were carried out according to the guidelines established by the Brazilian College for Animal Experimentation (COBEA).

Conclusions
In summary, four sub-extracts from D. alata have been examined for their protective activity against the neuromuscular blockade caused by Bothrops jararacussu venom. The active CH 2 Cl 2 extract has been extensively analyzed for its chemical constituents, resulting in the isolation of four lupane-type triterpenoids, nine isoflavonoids, one chalcone, one aurone and three phenolic compounds. On the basis of the present study, further pharmacological evaluation of each isolated compound will be need to ascertain which of them are responsible for the activity, as well as for establishing the mechanism of action.