New Octadecanoid Enantiomers from the Whole Plants of Plantago depressa

In this study, 19 octadecanoid derivatives—four pairs of enantiomers (1–8), two racemic/scalemic mixtures (9–10), and nine biosynthetically related analogues—were obtained from the ethanolic extract of a Chinese medicinal plant, Plantago depressa Willd. Their structures were elucidated on the basis of detailed spectroscopic analyses, with the absolute configurations of the new compounds assigned by time-dependent density functional theory (TD-DFT)-based electronic circular dichroism (ECD) calculations. Six of them (1, 3–6, and 9) were reported for the first time, while 2, 7, and 8 have been previously described as derivatives and are currently obtained as natural products. Our bioassays have established that selective compounds show in vitro anti-inflammatory activity by inhibiting lipopolysaccharide-induced nitric oxide (NO) production in mouse macrophage RAW 264.7 cells.


Introduction
The genus Plantago L. (family Plantaginaceae) consists of more than 190 species that are widely distributed in temperate and tropical areas all over the world. There are 20 Plantago plants that grow in China, including two invasive and one cultivated species [1]. P. depressa Willd. is a very common species found in most Asian countries [1], and its whole plants have long been used in traditional Chinese medicine as "Cheqian Cao" for the treatment of oedema, cough, carbuncle, etc. [2]. Previous chemical investigations of this medicinal plant have revealed the presence of phenylethanoid glycosides [3][4][5], iridoid glucosides [6,7], alkaloids [8], and so on [6,7,9]. However, few reports have dealt with the lipid constituents from P. depressa until now [10]. In the present work, we carried out an intensive chemical study on the EtOAc partition generated from the ethanolic extract of the whole plants of P. depressa, which resulted in the isolation of a series of fatty acid derivatives-four pairs of enantiomers (1-8), two racemic/scalemic mixtures (9-10) (Figure 1), and nine related analogues (11)(12)(13)(14)(15)(16)(17)(18)(19). The structures of these compounds were fully characterized by comprehensive spectroscopic analyses, with the absolute stereochemistry of the new compounds established via calculated ECD data. The in vitro antimicrobial, anti-acetylcholinesterase, and anti-inflammatory activities of these lipid molecules were tested; only two known compounds exhibited moderate anti-inflammatory effects. Herein, we describe the separation, structural characterization, and biological evaluations of these plant lipids.
anti-inflammatory effects. Herein, we describe the separation, structural characterization, and biological evaluations of these plant lipids.
We therefore employed the time-dependent density functional theory (TD-DFT) method to calculate the ECD spectra ( Figure 3) of the two enantiomers and finally differentiated them from each other.    Compounds 3/4 had the molecular formula of C20H32O4 as deduced from the (+)-HR-ESIMS ion peak at m/z 337.2369 ([M + H] + , calcd 337.2373), which was 14 mass units (CH2) more than that of compounds 7/8 [14,15] indicative of a methylated analogue. Analysis of the NMR data (Tables 1 and  2) for compounds 3/4 confirmed this hypothesis, with extra signals for a methoxy group (δH 3.28, δC 56.7) and the downfield shifted C-16 resonance (δC 84.5) in contrast with that (δC 74.2) in compounds 7/8. Further inspection of 2D NMR data ( Figure 2) corroborated this structural assignment, revealing key HMBC correlations with the methoxy protons to C-16. Compounds 3/4 were thus characterized to be methyl (10E,12E,14E)-16-methoxy-9-oxo-10,12,14-octadecatrienoate. Similar to compounds 1/2, the optical rotation and ECD data of compounds 3/4 suggested a scalemic mixture with nearly zero [α] D and no Cotton effect, respectively. The two pure enantiomers were further separated from each other by chiral HPLC and structurally differentiated by comparing their experimental ECD spectra with the calculated ones ( Figure 3).   Compounds 3/4 had the molecular formula of C20H32O4 as deduced from the (+)-HR-ESIMS ion peak at m/z 337.2369 ([M + H] + , calcd 337.2373), which was 14 mass units (CH2) more than that of compounds 7/8 [14,15] indicative of a methylated analogue. Analysis of the NMR data (Tables 1 and  2) for compounds 3/4 confirmed this hypothesis, with extra signals for a methoxy group (δH 3.28, δC 56.7) and the downfield shifted C-16 resonance (δC 84.5) in contrast with that (δC 74.2) in compounds 7/8. Further inspection of 2D NMR data ( Figure 2) corroborated this structural assignment, revealing key HMBC correlations with the methoxy protons to C-16. Compounds 3/4 were thus characterized to be methyl (10E,12E,14E)-16-methoxy-9-oxo-10,12,14-octadecatrienoate. Similar to compounds 1/2, the optical rotation and ECD data of compounds 3/4 suggested a scalemic mixture with nearly zero [α] D and no Cotton effect, respectively. The two pure enantiomers were further separated from each other by chiral HPLC and structurally differentiated by comparing their experimental ECD spectra with the calculated ones ( Figure 3).    [14,15] indicative of a methylated analogue. Analysis of the NMR data (Tables 1 and 2) for compounds 3/4 confirmed this hypothesis, with extra signals for a methoxy group (δ H 3.28, δ C 56.7) and the downfield shifted C-16 resonance (δ C 84.5) in contrast with that (δ C 74.2) in compounds 7/8. Further inspection of 2D NMR data ( Figure 2) corroborated this structural assignment, revealing key HMBC correlations with the methoxy protons to C-16. Compounds 3/4 were thus characterized to be methyl (10E,12E,14E)-16-methoxy-9-oxo-10,12,14-octadecatrienoate. Similar to compounds 1/2, the optical rotation and ECD data of compounds 3/4 suggested a scalemic mixture with nearly zero [α] D and no Cotton effect, respectively. The two pure enantiomers were further separated from each other by chiral HPLC and structurally differentiated by comparing their experimental ECD spectra with the calculated ones ( Figure 3).
Compounds 5/6 were determined to be monochlorinated on the basis of the ESIMS (electrospray ionization mass spectrometry) isotope ion peak at m/z 397.  (Tables 1 and 2) for compounds 5/6 also displayed resonances for several functional groups as those in compounds 1/2, such as two carbonyls (δ C 174.5 and 201.0), a diene (δ C 130.6, 141.1 and 130.3, 141.7; δ H 6.20, 7.15 and 6.49, 6.21), and an ester methoxyl (δ C 51.7; δ H 3.67). Meanwhile, compounds 5/6 possessed three sp 3 methines (δ C 64.6, 71.0 and 76.5; δ H 3.97, 4.70 and 3.62) compared with only one in compounds 1/2, which was ascribed to two hydroxyl and a chlorine substituents by analyzing the molecular composition and chemical shifts of these methines. Subsequent acquisition of 2D 1 H-1 H COSY and HMBC data (Figure 2) confirmed the establishment of the planar structure of compounds 5/6 as shown, and the substitution pattern of 14-OH, 15-OH, and 16-Cl for the C-14-C-16 fragment was supported by the lower chemical shift for C-16 than those for C-14 and C-15 [16]. The relative configuration of compounds 5/6 was determined by the J-based configuration analysis method [17]. The magnitudes of J 14,15 (2.7 Hz) and J 15,16 (7.6 Hz) indicated a syn-relationship between H-14 and H-15 and an anti-relationship between H-15 and H-16, respectively. Alerted by the cases of compounds 1-4, compounds 5/6 were also subjected to chiral HPLC analysis and indeed proved to be another pair of enantiomers. The absolute configurations of compounds 5/6 were further assigned by comparing their experimental ECD spectra with the computed ones ( Figure 3).
Compound 9 was assigned the molecular formula of C 19 H 32 O 5 Cl-same as compounds 5/6-based on the (+)-HR-ESIMS ion peak at m/z 397.1757 ([M + Na] + , calcd 397.1752), supportive of an isomer of the latter. Analysis of the NMR data (Tables 1 and 2) for compound 9 corroborated this conclusion, with very similar NMR data suggesting that they were diastereoisomers of the same planar structure; this was further confirmed by examination of 2D NMR correlations (Figure 2). Detailed NMR comparison between compounds 9 and 5/6 revealed nearly superimposable 1 H and 13 C NMR spectra, and the only difference was attributable to signals across CH-13 to CH-16 moiety. Obvious NMR variations were observed for resonances from H-13 to H-16, C-13 to C-14, and J 14,15 (Tables 1  and 2), which all suggested an inverted C-14 configuration in compound 9 compared with that in compounds 5/6. The structure and relative configuration of compound 9 were thus elucidated. It was inferable that compound 9 could also be enantiomeric mixture in light of the aforementioned examples of its cometabolites. However, it was not further separated on chiral HPLC due to degradation during storage, as indicated by subsequent 1 H NMR measurement. Moreover, the scarce amount of sample prevented us from further investigation.
Most compounds (only those with enough amount) were screened for their antimicrobial, anti-acetylcholinesterase, and anti-inflammatory activities (Supplementary Information Tables S1 and S2); only compounds 13 and 18 displayed anti-inflammatory effect with moderate inhibition against nitric oxide (NO) production with IC 50 values of 13.08 ± 0.25 and 7.64 ± 0.21 µM, respectively.

Plant Material
The whole plants of P. depressa were collected in June 2016 at Mount Kunyu, Shandong Province, and were authenticated by Prof. Jie Zhou from University of Jinan. A voucher specimen has been deposited at School of Biological Science and Technology, University of Jinan (Accession number: npmc-007).

Antimicrobial Assay
The antimicrobial assays were performed as we have reported earlier [25].

Anti-Acetylcholinesterase Assay
The anti-acetylcholinesterase assay was conducted as we have described earlier [26].

Anti-Inflammatory Assay
Determination of nitric oxide production. Briefly, RAW 264.7 cells were plated into 96-well plates and pretreated with a series of concentrations of compounds for 1 h before treatment with 1 µg/mL LPS. After 24 h incubation, detection of accumulated nitric oxide in the cell supernatants was assayed by Griess reagent kit (Beyotime Institute of Biotechnology) according to the manufacturer's instructions. Equal volumes of culture supernatant and Griess reagent were mixed, and the absorbance at 540 nm was measured using a Microplate Reader (Tecan, Switzerland).
Cell viability assay. RAW 264.7 cells were seeded into 96-well plates at 1 × 10 4 cells/well and allowed to attach for 24 h. The medium was replaced with 100 µL medium containing the indicated concentrations of compounds and further incubated for 24 h. 10 µL of MTT (5 mg/mL in PBS) was added into each well and the plates were incubated for 4 h at 37 • C. Supernatants were aspirated and formed formazan was dissolved in 100 µL of dimethyl sulfoxide (DMSO). The optical density (OD) was measured at an absorbance wavelength of 490 nm using a Microplate Reader (Tecan, Switzerland).

ECD Calculations
Conformational analysis within an energy window of 3.0 kcal/mol was performed by using the OPLS3 [27,28]  The conformers were then further optimized with the software package Gaussian 09 [30] at the B3LYP/6-311++G(2d,p) level, and the harmonic vibrational frequencies were also calculated to confirm their stability. Then, the 30 lowest electronic transitions for the obtained conformers in vacuum were calculated using TD-DFT method at the B3LYP/6-311++G(2d,p) level. ECD spectra of the conformers were simulated using a Gaussian function with a half-bandwidth of 0.26 eV. The overall theoretical ECD spectra were obtained according to the Boltzmann weighting of each conformer.
Supplementary Materials: The following materials are available online, raw spectroscopic data including chiral HPLC analyses, HR-ESIMS, and NMR ( 1 H, 13