12α-hydroxy-N-demethyl-sauroxine, a lycodane type alkaloid from Phlegmariurus saururus

Abstract 12α-hydroxy-N-demethyl-sauroxine (1), another new Lycopodium alkaloid from the Lycodane group, was isolated from Phlegmariurus saururus (Lam.) B. Øllg. (Lycopodiaceae). Elucidation of the chemical structure and relative stereochemistry were stated by spectroscopic data and chemical correlation. In addition, the inhibitory activity on acetylcholinesterase for 1 was determined as well as for N-methyllycodine (2), a derivative with the same nucleus, previously identified in P. saururus (IC50 = 33.8 ± 0.8 μM and 547.5 ± 0.5 μM, respectively) and N-demethylsauroxine (3) whose inhibition in the actual conditions was better than the previously informed. Graphical Abstract


Introduction
Lycopodium alkaloids are characterized by their unusual structures, intriguing heterocycles with diverse moieties Wang et al. 2016;Hirosawa et al. 2009); some of them have notable pharmacological effects (Olafsdottir et al. 2013). This class of alkaloids only occur in the Lycopodiaceae P. Beauv. ex Mirb. family. Belonging to this family, Phlegmariurus saururus (Lam.) B. Øllg. (ex-Huperzia saururus (Lam.) Trevis.; Lycopodium saururus Lam.) is a native species used in Argentina as a memory improver (Mart ınez Crovetto R 1981). The alkaloid extract (AE) of the species provokes inhibitory action on acetylcholinesterase (AChE) enzyme (Ortega et al. 2004a) and positive modulatory effect on memory and learning (Ortega et al. 2006;Vallejo et al. 2007). Looking for new chemical structures and especially those biologically active in relation to learning and memory, we have identified ten Lycopodium alkaloids in the species (Ayer et al. 1965;Ortega et al. 2004aOrtega et al. , 2004bVallejo et al. 2013). Among others, we isolated sauroxine (see Figure S1), that produces in vitro inhibitory effect on the AChE (Puiatti et al. 2013), and sauroine, an ex vivo and in vivo mnemonic phenomenon enhancer (Vallejo et al. 2009). Lycopodium alkaloids from Lycodane skeleton, like sauroxine, have been reported to have modulation on that enzyme activity (Halldorsdottir et al. 2015).
Continuing with the search for new structures, the objective of the present investigation was to isolate and identify another Dragendorff positive compound from the AE. Thus, a Lycodane-type alkaloid was isolated from aerial parts of P. saururus, 12ahydroxy-N-demethyl-sauroxine (1) (Figure 1).
Supporting the previous description for 1, it is also possible to observe the influence of the hydroxyl group on the neighbour carbons. Compared to the previous reported alkaloids, in sauroxine, C-7, C-10, C-11 and C-12 were detected at d C ¼33.8, 18.6, 24.9 and 32.6, respectively. In N-demethyl-sauroxine, they were present at d C ¼32.7, 22.4, 22.8 and 40.1, while in 1 the four carbon signals are displaced to a lower field as it can be seen in Table S1.
NMR 2D spectra were analysed and connections between C-2 to C-3, C-6 to C-7 and C-8, C-9 to C-11, and C-14 to C-16 were identified by 1 H-1 H COSY spectrum ( Figure S2). The a-pyridone ring (A) presence was corroborated by the HMBC correlations ( Figure S2). H-2 showed correlations with C-1, C-3 and C-4, H-3 with C-1, C-4 and C-5. Connection between A and B rings is showed by correlation of H-6 with C-4. B ring was constituted by connections of H-6 with C-7, C-12 and C-13 and H-7 to the same carbon atoms. Linkage with D ring was demonstrated by correlation in H-8 with C-7, C-12 and C-15. D ring was confirmed by connections between H-14 with C-4, C-8, C-12 and C-15, H-15 with C-8, C-12 and C-16; and H-16 with C-15. Finally, correlations in H-9 with C-10, and C-11, and H-11 with C-7 and C-13 allowed the conformation of C ring.
Stereochemistry of 1 was determined by NOESY spectrum. Spatial correlations were detected for H-3a and H-6b; H-10b and H-14a; H-14b with H-3b and H-16; H-7 with H-11b as well ( Figure S3). No correlations present in 1 were observed between H-12 to H-6b and H-3a, confirming the absence of H in C-12. Thus, hydroxyl group at C-12 is a-oriented, so D ring is cis oriented to A. The same configuration is present in sauroxine and N-demethyl-sauroxine, being unique in the three Lycodane type alkaloids ). Hence, a common biosynthetic pathway for these structures is suggested.
Three P. saururus alkaloids have been assessed in the present work as per their AChE inhibitory activity. The first one (1), as chemically demonstrated, is a new Lycopodium alkaloid, which showed good inhibition (IC 50 ¼ 33.8 ± 0.8 lM).
On the other hand, activity of alkaloid 2 (previously isolated by us) is informed for the first time, and it resulted a moderate inhibitor (IC 50 ¼ 547.5 ± 0.5 lM). Finally, the third alkaloid also previously descripted by us, showed a remarkable activity (IC 50 ¼ 1.38 ± 0.5 lM). Although the inhibitory activity of 3 had already been evaluated when it was isolated, in this opportunity the three components were assayed by using a more sensitive methodology than that being previously assessed for 3. Thus, the difference in its IC 50 result compared to that published in 2013 .
Analysing the inhibitory effect on the AChE of the alkaloids isolates, N-demethylsauroxine (3) has the higher level of inhibition (IC 50 ¼ 1.38 ± 0.5 lM) (Vallejo et al. 2007). If chemical structure of 3 is compared to 1 (IC 50 ¼ 33.8 ± 0.8 lM), we could infer that the presence of a hydroxyl group at C-12 diminishes its effect. On the other hand, it seems also that aromatic ring C in 2 without the presence of a carbonyl group, drastically affects its inhibitory effect on AChE (IC 50 ¼ 547.5 ± 0.5 lM). Thus, a less polar and planar moiety could be interacting with lower efficiency with the enzyme.

General experimental procedures
A Jasco P-1010 polarimeter (Jasco, Tokyo, Japan) was used to determine the optical rotation. IR spectrum was recorded in a Nicolet 5SXC spectrophotometer (Thermo Fisher Scientific, Massachusetts, USA). NMR spectroscopy was developed in a Bruker Advance NMR 400 spectrometer (Bruker Co, Karlsruhe, Germany), at 400 MHz ( 1 H) and 100 MHz ( 13 C), using Cl 3 CD as solvent. HRESIMS analysis was performed in a Bruker micrOTOF-Q II spectrometer (Bruker Co, Karlsruhe, Germany), in MeOH. Column chromatography was developed using Merck Sephadex LH-20 filtration gel, and purification was made by TLC using Merck silica gel 60F 254 plates. In the latter, components were detected by UV lamp as dark spots. The acetylcholinesterase inhibitory assay was performed in a Tecan ELISA microplate reader (BioTek, Winooski, USA), in 96-well microplates. Acetylthiocholine iodide (ACTI) and 5,5-dithio-bis(2-ni-trobenzoic acid) (DTNB) were purchased from Sigma-Aldrich (USA).

Plant material
P. saururus was collected in November 2016, in Pampa de Achala, San Alberto Department, Province of C ordoba, Argentina. A voucher specimen is deposited at the Museo Bot anico de C ordoba herbarium (CORD) as CORD 684.

Extraction and isolation
First, dried aerial parts (2.1 kg) were moistened with NaOH 0.1 M and extracted with Cl 3 CH using a Soxhlet equipment. Dried extract obtained (65g) was solubilized in HCl 0.1 N and partitioned with n-hexane to remove impurities. Secondly, acid aqueous extract was alkalinized with NaOH 1 M until pH 12 and partitioned with Cl 3 CH by using a liquid-liquid extractor. Consequently, 1.33 g of extract (E1) was obtained. E1 was subjected to Sephadex LH-20 column chromatography using Cl 3 CH/MeOH (1:1) as a mobile phase and two fractions were generated. Fraction 2 was purified by Sephadex LH-20 with acetone as the mobile phase, giving fifteen fractions (A-O). Fraction E was purified by TLC using cyclohexane/diethylamine (1:1) as mobile phase obtaining seven fractions; E3 yielded 1.6 mg of 1. Fraction B was purified also by TLC but using cyclohexane/Cl 3 CH/diethylamine (5:4:1); four fractions were obtained and B4 yielded 0.9 mg of the already described N-methyllycodine alkaloid (2) previously identified in P. saururus (Ortega et al. 2004a). 3 was isolated as we reported before , yielding 1.6 mg.

Acetylcholinesterase inhibition assay
Activity of AChE was determined according to the modified Ellman et al. method; the membranes of red blood cells were obtained from human blood (Ellman et al. 1961). Briefly, in a 96-well microplate 0.21mM ATCI (30 mL) in distillated water was added to a mixture containing 0.21mM DTNB (200 mL) in phosphate buffer pH 7.4, the enzyme (5 mL), and phosphate buffer pH 7.4, when normal activity was measured, as well as, for the alkaloids solutions (30 mL). Several dilutions of alkaloids as hydrochloride were prepared from stock arriving at every final concentration evaluated. These solutions of 1, 2 and 3 were used to determine their IC 50 . Physostigmine salicylate was used as positive control. Determinations were carried out by triplicating at 405 nm, after 30 s cycle, for 3 minutes and then, the reaction mixtures were incubated at room temperature for 30 min. After lapse, the absorbances were recorded again in identical way. Inhibition for positive control and alkaloids solutions were expressed as a percentage of the activity related to control (normal activity without an inhibitor). Results were analysed by GraphPad Prism 6 and thus, the IC 50 obtained.