Artificial Triterpenoid Fatty Acid Ester Isolated From the Leaves of Phytolacca icosandra L

Abstract The methanol extract form the leaves of Phytolacca icosandra L., afforded the unprecedented artificial triterpenoid fatty acid ester 1 derived from the new natural triterpenoid phytolaccagenic acid 3-O-myristate (1a), along with the three known triterpenoids serjanic, acinosolic and phytolaccagenic acid (2 – 4). Their structures were stablished by HR-EI-MS, 1D and 2D NMR techniques. The possible mechanistic formation of 1 is proposed, and the in vitro toxicity of all compounds was assessed using the brine shrimp lethality assay (BSLA). Graphic Abstract Electronic supplementary material The online version of this article (10.1007/s13659-020-00249-x) contains supplementary material, which is available to authorized users.


Introduction
The chemistry of Phytolacca ssp. is fairly wide and comprises a variety of secondary metabolites, composed mainly by triterpenoids, flavonoids and lignans [1]. Plants belonging to genus Phytolacca have been used in folk medicine for the treatment of several affections such as edema, rheumatism and dermatitis [1][2][3]; also as a molluscicidal plant in schistosomiasis prevention and control [4,5]. Several studies on P. icosandra have reported its antisecretory, anthelmintic, ovicidal and larvicidal activity [6][7][8].
Phytochemical analysis of P. icosandra has lead to the isolation of several serjanic and spergulagenic acids [9,10] and a previous investigation of the fruits yielded a novel peltogynoid, together with triterpenoids, neo-lignanes and 6′palmitoyl-α-d-glucoside sterols [11]. As part of our continuing search for new bioactive constituents from plants of the Phytolacca genus, the methanolic extract of the leaves of P. icosandra was investigated. As a result, a new artificial triterpenoid fatty acid ester (1) was isolated along with three other known pentacyclic triterpenoids 2-4 ( Fig. 1). We also comment on the possible formation of 1, and the in vitro toxicity of all compounds against brine shrimps (Artemia salina).

Structure Elucidation of Isolated Compounds
Compound 1 was isolated as a white wax. A molecular formula of C 48 H 80 O 8 was assigned from its HR-MS spectra, which showed a molecuar ion peak [M] + at m/z 784.5859 (calcd. 784.5853) accounting for nine degrees of unsaturation. The IR spectrum exhibited absorption bands due to the presence of hydroxyl and carboxylic acid groups (2900-3400 cm −1 ), carbonyl groups (1701-1705 cm −1 ), olefinic bond (1472 cm −1 ) and long chain alkanes band (728 cm −1 ). The 1 H NMR spectrum in conjunction with the HMQC spectrum, revealed the presence of five tertiary methyl groups at δ H/C 0.69/16. The 13 C NMR spectra showed that compound 1 has an 3,23-dihydroxy-olean-12(13)-en-28,30-dioic acid-30-methyl ester triterpene skeleton, because in addition to the presence of the peaks assigned to the five tertiary methyl mentioned above, it is possible to locate peaks corresponding to six sp 3 quaternary carbons [δ C 36. (C-12), 142.8 (C-13)], and two carbonyl carbons [δ C 176 (C-30), 183.4 (C-28) ppm]. An acyclic acetonide moiety, previously described as a 2,2-dioxy-propane group, esterified by the aforementionated long chain fatty acid were also elucidated in the molecule, across the 13 C NMR spectra.
Although few, there has been some reports on natural occurring triterpenoidal acetonides from plants [17][18][19]. Despite the fact that 1 is considered unusual being an acyclic acetonide ketal, it is assumed to be an artifact derived from phytolaccagenic acid 3β-O-myristate (1a) during the chromatographic process, in which acetone was used as solvent [20,21]. The proposed mechanism in the formation of 1 from 1a involves firstly a nucleophilic attack of the C-23 hydroxyl to a protonated acetone molecule, followed by an intramolecular nucleophilic substitution at the fatty acid carbonyl at C-3 (Fig. 2).
Finally, the toxicity of all compounds was assayed in the brine shrimp lethality assay [22] and compound 1 exhibited mild toxicity against Artemia; results are shown in Table 1.

General Procedures
Optical rotation was measured in Karl-ZEISS, Model 93,772 equipment. IR spectra were obtained from KBr pellets with Shimadzu IR-408 equipment. Solution 1 H, 13 C{ 1 H}, 1 H, 1 H-COSY, HSQC, HMBC and NOESY NMR spectra were recorded on Bruker Avance 600 MHz spectrometer at Laboratorio Nacional de Resonancia Magnética Nuclear, Instituto Venezolano de Investigaciones Cientificas (IVIC), using CDCl 3 and MeOD as solvent. Peak positions are relative to tetramethylsilane for 1 H and 13 C{ 1 H}. The chemical shifts (δ) were measured according to IUPAC [23], expressed in parts per million (ppm) and were calibrated against the residual solvent resonance ( 1 H) or the deuterated solvent triplet ( 13 C). Coupling constants J are given in Hertz (Hz) as absolute values. The multiplicity of the signals are indicated as s, d, t, q, or m for singlets, doublets, triplets, quartets or multiplets respectively. All NMR spectra were recorded at room temperature (25 °C) in CDCl 3 dried over molecular sieves. ESI-MS was run on a TSQ QUANTUM, Ultra AM, ThermoScientific Spectro-photometer and the HR-EI-MS analysis

Identification of Known Compounds
Known compounds were identified by comparison of their physical constants and NMR spectroscopic data with those reported in the literature [24][25][26].

Brine Shrimp Lethality Assay
The assay was performed as described previously by Meyer et al. [22] with some minor modifications. Brine shrimp eggs (Gulf Breeze®) were hatched in artificial sea water prepared with commercial salt mixture (Instant Ocean®), illuminated and oxygenated with an aquarium pump. After 48 h incubation at 27 °C, 10 shrimps were transferred with a Pasteur pipette to three sample vials for each of three doses (100, 50, 10 μg/mL) for a total of nine vials. The sample was prepared by dissolving the compound 1 (3 mg) in CHCl 3 (5 mL) and transferring the solution to each vial (833, 417 or 83 μL solution for 100, 50 or 10 ppm doses) followed by high vacuum for 1 h. After the solvent was evaporated, the compound was redissolved in 20 μL of Tween 80® and 5 mL of artificial sea water were added to achieve the correct concentration. Survivors were counted and the percent deaths at each dose and control were determined. Tween 80® at this concentration did not affect this bioassay. The LC 50 and 95% confidence intervals were calculated from 24 h counts, using the Probit analysis method [27].