Isolation and Identification of Phytocompounds from Maytenus dhofarensis and Their Biological Potentials

Maytenus dhofarensis Sebsebe (Celestraceae) is a naturally growing shrub in Oman. It is not a reputed medicinal plant in Oman, but it is regionally endemic and causes shivering attacks on goats that graze on it. The chemical investigation of the hexane and chloroform extracts of the fruits and stems of M. dhofarensis afforded dihydro-β-agarofuran-type sesquiterpene pyridine alkaloid (1), lupanyl myristoate (2) and lignanolactone (3). Compounds (1–3) are new isolates from M. dhofarensis. The structures of these compounds were assigned through comprehensive IR, NMR, and ESI-MS analyses, and the relative configurations of compounds 1 and 3 were deduced from density function theory (DFT) calculations and NMR experiments. Compound 1 was assayed against the kinase enzyme and showed no inhibition activity for p38 alpha and delta at a 10 µM test concentration. Compound 3 inhibited the 2,2′-diphenyl-1-picrylhydrazyl radical (DPPH) by 69.5%, compared to 70.9% and 78.0% for gallic acid and butylated hydroxyanisole, respectively, which were used as positive controls.

Maytenus dhofarensis is a regionally endemic spiny shrub found growing naturally in the Dhofar region of Oman [38]. It is not a reputed medicinal plant in Oman but causes shivering attacks on goats that graze on it. To date, no reports on its secondary metabolites have previously been described. The lack of elaborate phytochemical and pharmacological work on this species stimulated our interest in examining the plant for structurally novel compounds and bioactivity. Herein, we report the isolation and structural characterization of a new dihydro-β-agarofuran-type sesquiterpene pyridine alkaloid (1), lupanyl myristoate (2), and lignanolactone (3) (Figure 1) from the fruits and stems of the plant, and the bioassay results of the more-abundant isolates for their antioxidant and kinase-inhibitory activities. Their structures were elucidated from the interpretation of spectral data and the relative and the bioassay results of the more-abundant isolates for their antioxidant and kinaseinhibitory activities. Their structures were elucidated from the interpretation of spectral data and the relative configurations of compounds 1 and 3 were determined from observed and calculated chemical shift values for their diastereoisomers using DP4+ probability [39,40] analysis. Compound 3 showed 2,2′-diphenyl-1-picrylhydrazyl radical (DPPH)-scavenging activities.
The relative configuration of 1 was established by the 2D NOESY spectrum and by comparing it with previous studies [16,18]. DFT-predicted chemical shifts for 9R and 9S diastereomers were compared with the observed values for the isolated compound, and the best match using the DP4+ probability [39] favored the 9R configuration for compound 1. The NOESY spectrum also did not show any correlation between H-1 and H-9, supporting the 9R configuration assignment. Predicted chemical shift values are listed in Table S1, while the results of the DP4+ analyses are summarized in Table S2. Based on the above assignments, the structure of compound 1 was identified as a dihydro-β-agarofuran-type sesquiterpene pyridine alkaloid, which was named Maytendhofarene (Figure 1).
Compound 1 was investigated for kinase-inhibition activity using the homogenous time-resolved fluorescence (HTRF) detection kit [41] but showed no inhibition activity for p38 alpha and delta at a 10 µM test concentration ( Figure 2). p38 MAP kinases have been implicated in a wide range of complex biological processes, such as cell differentiation and proliferation, cell death, cell migration, and invasion [42]. The dysregulation of p38 MAPK is associated with diverse diseases such as chronic inflammation and cancer and can act as a tumor suppressor or tumor inducer [43].
OH -3.10 5, 7, 14 The relative configuration of 1 was established by the 2D NOESY spectrum and by comparing it with previous studies [16,18]. DFT-predicted chemical shifts for 9R and 9S diastereomers were compared with the observed values for the isolated compound, and the best match using the DP4+ probability [39] favored the 9R configuration for compound 1. The NOESY spectrum also did not show any correlation between H-1 and H-9, supporting the 9R configuration assignment. Predicted chemical shift values are listed in Table S1, while the results of the DP4+ analyses are summarized in Table S2. Based on the above assignments, the structure of compound 1 was identified as a dihydro-β-agarofuran-type sesquiterpene pyridine alkaloid, which was named Maytendhofarene (Figure 1).
Compound 1 was investigated for kinase-inhibition activity using the homogenous time-resolved fluorescence (HTRF) detection kit [41] but showed no inhibition activity for p38 alpha and delta at a 10 µM test concentration (Figure 2). p38 MAP kinases have been implicated in a wide range of complex biological processes, such as cell differentiation and proliferation, cell death, cell migration, and invasion [42]. The dysregulation of p38 MAPK is associated with diverse diseases such as chronic inflammation and cancer and can act as a tumor suppressor or tumor inducer [43].  Compound 2, whose molecular formula was identified as C 44 (Table 3) showed signals for six aromatic protons and some phenolic hydroxy, two methoxy, one methane, and three methylene groups for compound 3. It also displayed isochronous signals at δ H 4.04 and 3.99, δ H 3.10, and 2.91, and δ H 2.59 and 2.49 for oxymethylene, acetoxymethylene, and benzylmethylene protons ( Table 3). All the methylene protons in compound 3 are thus diastereotopic atoms. A methine proton at δ H 2.52 (H-8 ) showed spin-spin couplings to the oxymethylene protons (H-9 a and H-9 b) and benzylmethylene protons (H-7 a and H-7 b). The diastereotopic acetoxy protons were only coupled to each other in the 1 H-1 H COSY map and must be flanked by quaternary carbons. The interpretation of the IR, 1 H NMR chemical shifts, and coupling patterns suggested a β-hydroxy-β-phenyl-γ-benzyl δ-valerolactone substructure for compound 3. The two methoxy groups resonated as singlets at δ H 3.64 and 3.85, and the six aromatic protons constituted two separate ABX coupling systems with characteristic splitting patterns of aromatic protons in 1,3,4 relative positions. Each of the four aromatic protons resonated as a doublet (J = 7.9 Hz) at δ H 6.84, 6.82, 6.63, and 6.62, and the other aromatic protons as broad singlets at δ H 6.69 and 6.60, respectively. The aromatic signals and spin-spin couplings are consistent with the presence of two 3, 4-di-substituted phenyl units in compound 3. The 13 C NMR Broad Band and DEPT experiments resolved twenty carbon resonances for compound 3, which included one carbonyl at δ C 179.1 (C-9) and two tri-substituted aromatic units ( Table 3). The spectra also displayed the following carbon signals: one oxymethylene at δ C 70.7 (C-9 ), one acetoxy methylene at δ C 42.4 (C-8), one benzylmethylene at δ C 31.9 (C-7 ), one methine at δ C 44.2 (C-8 ), one oxygenated quaternary carbon at δ C 76.9 (C-7), and signals for two methoxy groups at δ C 56.4 and 56.3, respectively. The assignment of methoxy groups to C-4 and C-4 was complemented by connectivity between δ H 3.85 and δ C 144.  (Table S3) with the observed values for the isolated compound, and the best match (98.9%) using the DP4 probability [40] favored H-7S, H-8 R configuration for compound 3. Compound 3 demonstrated radical scavenging property, inhibiting [45] DPPH by 69.5%, compared to 70.9% and 78.0% for gallic acid and butylated hydroxyanisole, which were used as controls.

General Experimental Procedures
IR spectra were obtained with a Nicolet FT-IR spectrometer. 1 H and 13 C NMR spectra were recorded in CDCl 3 with Bruker Advance NMR spectrometer operating at 700 MH with TMS as the internal standard. ESIMS was recorded on a Quattro Ultima Platinum Tandem quadrupole mass spectrometer (Micromass, Wilmslow, UK). ESIMS data were acquired on an Agilent 6400 Triple Quad LC/MS and using HRESIMS. The column chromatography (CC) was performed using EM Science Silica gel 60 (70-230 mesh ASTM). Whatman precoated silica-gel (60A K6F) analytical plates (20 × 20 cm) were used for TLC, with compounds visualized by a UV lamp and spraying with 10% (v/v) H 2 SO 4 or Molybdophosphoric acid-isopropanol followed by heating. All absorbance measurements were recorded using a Shimadzu UV spectrophotometer.

Extraction and Isolation
The fruits of M. dhofarensis were dried in a hot room (at 42 • C) for 3 weeks in the Department of Chemistry, Sultan Qaboos University, and the seeds were separated from the calyx and milled to give 576 g of powdered seed. The seeds (288 g) were extracted with chloroform (2 × 1800 mL) by maceration at room temperature for three days each and concentrated under vacuum at 25-30 • C to give a gummy residue (208.6 g). The column chromatography of a portion of the chloroform extract (44.75 g) on silica gel (895 g), using gradient mixtures of n-hexane-CHCl 3 , CHCl 3, and CHCl 3 -EtOH as eluent, gave a variable number of fractions, which were combined based on their TLC profiles.
A portion of the dried and powdered calyx (275 g) of M. dhofarensis was extracted with hexane by Soxhlet (2 × 600 mL) for four hours. The solvent was removed in a vacuum to yield a hexane extract (11.5 g). The separation of the extract was undertaken using column chromatography with silica gel (120 g) and using hexane and gradient mixtures of hexane-ethyl acetate with a collection of 100 mL fractions. TLC analysis using hexane: EtOAc

Computational Studies of Compounds 1 and 3
Density functional theory (DFT) was used to predict the NMR chemical shifts for the diastereomers of compound 1 (9R and 9S) and compound 3, (7R-8 R and 7S-8 R). For each, conformers up to an upper limit of 10 kcal/mol were identified using the MMFF force field as implemented by MarvinView software (version 17.2.6.0) [47]. The selected rotamers were optimized using Gaussian (G09W) software (Gaussian 09, revision E.01) [40] at the B3LYP/6-31G+(d,p) (gas-phase) level of theory, and subsequent frequency calculations confirmed the absence of any imaginary frequencies in the minimized structures. Isotropic shielding constants were calculated at the mPW1PW91/6-311+G(d,p)/PCM level using the GIAO method. The predicted values for each diastereomer were averaged using a Boltzmann weighting, and for each compound, the unscaled shielding constants were compared in the Bayesian-based DP4+ analyses utilizing the Excel file provided by the Sarotti group [39].

Homogenous Time-Resolved Fluorescence (HTRF) Kinase Assay
Compound 1 was diluted with a kinase buffer (with the p38 alpha enzyme or p38 delta enzyme) arising out of a stock solution of 10 mM in MDSO, giving a final concentration of 10 µM. Subsequently, 10 µL of the compound dilution was added to the enzyme, producing a final volume of 20 µL. Instead of a compound, 10 µL of KB was distributed in the wells of negative (NSB) and positive controls (STIM). The 96-well non-binding plate was centrifuged before a preincubation for 10 min at 37 • C, gently shaking at 150 rounds. After the preincubation, an ATP/ATF 2 solution was added to the wells and incubated again for 30 min at 37 • C. The assays were performed using the homogenous time-resolved fluorescence (HTRF) detection kit (Cisbio, Bedford, MA) by adding 10 µL of the HTRF detection solution (2.5 µL of PAb Anti-phospho ATF 2-Eu cryptate (1:400); 5 µL of MAb Anti GST-d2 (1:200); and 992.5 µL of HTRF detection buffer). The plate was incubated for 30 min at room temperature in the dark. The HTRF signal was read from the Victor Nivo ® and calculated as the ratio of signal from the 665 nm (acceptor) and 615 nm (donor) channels and multiplied by 10,000. The percent activity was calculated by normalizing the HTRF signal from each sample well to the mean HTRF signal from the DMSO-only control wells, using the following equation: Inhibition (%) =100 − ((ODsample − NSB)/(ODstim − NSB)) × 100% 3.5.2. Antioxidant Assay Activity Using 2,2 -Diphenyl-1-picrylhydrazyl (DPPH) Radical-Scavenging Method The free-radical scavenging activity of compound 3 was determined using the protocol reported by Morelli [45] with slight modifications: 8.65 mg of compound 3 was dissolved in CHCl 3 (1 mL), and an aliquot of this was diluted with CHCl 3 to give a solution of 0.4 mg/2.0 mL of compound 3. This solution was mixed with 2.0 mL of 100 µM DPPH solution prepared by dissolving 2 mg DPPH in 50 mL of 25% aqueous ethanol. Butyl hydroxyl anisole (BHA) and gallic acid solutions were used as positive controls, while 2.0 mL of 25% aqueous ethanol solution of 100 µM DPPH mixed with 2 mL of CHCl3 served as the blank solution. The absorbance at 517 nm of all prepared solutions was determined after 15 min of incubation in the dark against the blank solution.
The ability to scavenge DPPH was expressed as a percentage inhibition (% IP) of the DPPH radical.

Conclusions
Previously uninvestigated endemic M. dhofarensis yielded three new compounds (1-3), which are similar to the compounds found in the genus Mayteneus. However, compound 1 is structurally unique. It is a dihydro-β-agarofuran-type sesquiterpene pyridine alkaloid that differs structurally from the vast array of Celastraceous macrocyclic dihydro-β-agarofuran sesquiterpene pyridine alkaloids [48] due to the absence of a pyridine dicarboxylic acid macrocyclic bridge. This is structurally significant for an endemic plant of the genus Mayteneus. Compound 1 lacked kinase-inhibitory activity as hoped for in the original design of this work. It was obtained in the fruit and also detected in the alcohol extracts of the stem. The fruit is toxic to goats, and whenever a goat grazed the fruits, it fell ill with shivering attacks. Compound 3 is a new addition to the secondary metabolites from this genus. It was isolated from the stem and showed radical scavenging activity at a level comparable to gallic acid. This might help the plant to overcome oxidative stress.