Alkaloids from Hippeastrum papilio

Abstract

Galanthamine, an acetylcholinesterase inhibitor marketed as a hydrobromide salt (Razadyne®, Reminyl®) for the treatment of Alzheimer’s disease (AD), is obtained from Amaryllidaceae plants, especially those belonging to the genera Leucojum, Narcissus, Lycoris and Ungernia. The growing demand for galanthamine has prompted searches for new sources of this compound, as well as other bioactive alkaloids for the treatment of AD. In this paper we report the isolation of the new alkaloid 11β-hydroxygalanthamine, an epimer of the previously isolated alkaloid habranthine, which was identified using NMR techniques. It has been shown that 11β-hydroxygalanthamine has an important in vitro acetylcholinesterase inhibitory activity. Additionally, Hippeastrum papilio yielded substantial quantities of galanthamine.

1. Introduction

The object of many studies, Amaryllidaceae alkaloids contain a wide range of chemical structures and interesting biological properties [1], showing pronounced antimalarial [2], antitumoral [3] and acetylcholinesterase inhibitory activity [4]. The use of galanthamine in palliative therapy for mild-moderate AD [5] has prompted the search for analogous compounds bearing the galanthamine-type skeleton. Additionally, as most of the galanthamine used in clinics is supplied from natural sources, there is considerable interest in finding new Amaryllidaceae species for a sustainable production of galanthamine [4,6].

Plants of the genus Hippeastrum, which is endemic to South America, have yielded interesting bioactive compounds such as montanine, with significant psychopharmacological and acetylcholinesterase inhibitory activity [7,8] and candimine, active against Trychomonas vaginalis [9]. Recent nrDNA ITS sequences data have included it within the Hippeastroid subclade and alluded to a probable Brazilian origin [10].

We have identified six known alkaloids, including significant quantities of galanthamine which was the main alkaloid isolated, from the bulbs and leaves of Hippeastrum papilio (Ravenna) Van Scheepen, which grows in Southern Brazil. Furthermore, we have clarified the correct position of the hydroxyl-substituent in the alkaloid habranthine through the isolation of its epimer, the new alkaloid 11β-hydroxygalanthamine. Habranthine was isolated from Pancratium maritimum without certainty about the stereochemistry of the hydroxyl substituent at position 11 in the galanthamine-type skeleton. Using modern 2D NMR, we report the correct assignment of 11β-hydroxygalanthamine, a new alkaloid from Hippeastrum papilio, confirming that the previously reported habranthine is in fact 11α-hydroxygalanthamine. Furthermore, 11β-hydroxygalathamine has demonstrated a good ability to inhibit the acetylcholinesterase enzyme, with IC₅₀ of 14.5 ± 0.33 µM.

2. Results and Discussion

Bulbs and leaves of the plant showed similar alkaloid profiles by analytical TLC. After Vaccum Liquid Chromatography (VLC) and a purification process, the new alkaloid, 11β-hydroxygalanthamine (1), as well as six known alkaloids, namely galanthamine (2), which was found in relatively high quantities, narwedine (3), haemanthamine (4), 11-hydroxyvittatine (5), 8-O-demethylmaritidine (6) and vittatine (7), were isolated and identified by NMR, CD and MS spectrometry (Figure 1). Earlier studies have reported the isolation of galanthamine and other galanthamine-type alkaloids from Hippeastrum species, including European cultivars, but mainly as minor compounds.

Figure 1. Isolated alkaloids from Hippeastrum papilio.

Figure 1. Isolated alkaloids from Hippeastrum papilio.

Compound 1, 11β-hydroxygalanthamine, crystallized as white needles. EI-MS showed a molecular ion peak at m/z 303. The base peak at m/z 230 evidenced the loss of the hydroxyl group at C-3 and the C-11/C-12/NMe residue, in agreement with other galanthamine-type alkaloids [11]. The ¹H-NMR of compound 1 (Table 1) was very similar to that of habranthine previously isolated from Habranthus brachyandrum and Pancratium maritimum [12,13], with small differences in the chemical shifts of H-11 and H-1. The key to the assignation was the large coupling constant J_(11α,12β) = 10.8 Hz, observed in compound 1, which indicates a trans diaxial relationship between H-11α and H-12β, and therefore a β-position for the hydroxyl substituent. In contrast, the coupling constants observed for the H-11 of habranthine (J_(11β,12β) = 1.6 Hz) and (J_(11β,12α) = 4.5 Hz) indicated that the hydroxyl-substituent should be in the α-position (endo). Moreover, the H-12 protons in compound 1 are clearly separated in a double doublet, and 2D analysis confirmed the correct assignment, where NOESY correlations between H-12β and H-4/H-6β and between H-12α and NMe were observed. The complete assignment of the 11β-hydroxygalanthamine is presented in Table 1.

Alkaloid 1 has also proven to be an inhibitor of acetylcholinesterase like the majority of the galanthamine-type alkaloids. In an acetylcholinesterase inhibition screening of several Amaryllidaceae alkaloids, López et al. [14] found that habranthine, epimer of compound 1, showed similar activity to galanthamine. The β-configuration of the hydroxyl group at position 11 in 1 could be unfavourable for its interactions within the active site of the acetylcholinesterase enzyme. The IC₅₀ for 1 was 14.5 ± 0.33 µM, while galanthamine showed an IC₅₀ of 1.18 ± 0.07 µM.

Table 1. ¹H NMR, COSY, NOESY, HSQC and HMBC data of 11β-hydroxygalanthamine (400 MHz, CDCl₃).

Position	H δ	H δ (J in Hz)	COSY	NOESY	HSQC			HMBC
1	4.88	br s	H-2α, H-2β	H-2α, H-2β, H-11α		89.0	d	C-3, C-4a, C-11
2α	2.70	ddd (15.6, 5.2, 2.8)	H-1, H-2β	H-1, H-2β, H-3		32.3	t	C-1, C-3, C-4, C-10b
2β	2.37	br dt (15.6, 1.6)	H-1, H-2α, H-3	H-1, H-2α, H-3				--
3	4.20	br t (4.8)	H-2β, H-4	H-2α, H-2β, H-4		62.2	d	C-1, C-4, C-4a
4	6.29	dd (10.4, 4.8)	H-3, H-4a	H-3, H-4a		133.1	d	C-2, C-10b
4a	5.96	d (10.0)	H-4	H-4, H-6β, H-12β		122.6	d	C-1, C-3, C-10b
6α	3.60	d (15.2)	H-6β	H-6β, H-7, NMe		59.5	t	C-6a, C-7, C-10a, C-12, NMe
6β	3.93	d (14.8)	H-6α	H-4a, H-6α, H-12β				C-6a, C-7, C-10a, C-12, NMe
6a						129.9	s
7	6.61	d (8.0)	H-8	H-6α, H-8		122.4	d	C-6, C-9, C-10a
8	6.68	d (8.0)	H-7	H-7, OMe		111.8	d	C-6a, C-10
9						144.6	s
10						146.9	s
10a						128.9	s
10b						53.4	s
11α	4.05	dd (10.8, 4.0)	H-12α, H-12β	H-1, H-12α, H-12β, NMe		72.0	d	C-1, C-4a
12α	3.02	dd (14.0, 3.2)	H-11α, H-12β	H-11α, H-12β, NMe		62.1	t	C-6, C-10b, C-11
12β	3.17	dd (13.6, 10.8)	H-11α, H-12α	H-4a, H-6β, H-11α, H-12α				C-6, C-11, NMe
OMe	3.84	s (3H)		H-8		56.2	q	C-9
NMe	2.45	s (3H)		H-6α, H-11α, H-12α		43.2	q	C-6, C-12

Table 1. ¹H NMR, COSY, NOESY, HSQC and HMBC data of 11β-hydroxygalanthamine (400 MHz, CDCl₃).

Position	H δ	H δ (J in Hz)	COSY	NOESY	HSQC			HMBC
1	4.88	br s	H-2α, H-2β	H-2α, H-2β, H-11α		89.0	d	C-3, C-4a, C-11
2α	2.70	ddd (15.6, 5.2, 2.8)	H-1, H-2β	H-1, H-2β, H-3		32.3	t	C-1, C-3, C-4, C-10b
2β	2.37	br dt (15.6, 1.6)	H-1, H-2α, H-3	H-1, H-2α, H-3				--
3	4.20	br t (4.8)	H-2β, H-4	H-2α, H-2β, H-4		62.2	d	C-1, C-4, C-4a
4	6.29	dd (10.4, 4.8)	H-3, H-4a	H-3, H-4a		133.1	d	C-2, C-10b
4a	5.96	d (10.0)	H-4	H-4, H-6β, H-12β		122.6	d	C-1, C-3, C-10b
6α	3.60	d (15.2)	H-6β	H-6β, H-7, NMe		59.5	t	C-6a, C-7, C-10a, C-12, NMe
6β	3.93	d (14.8)	H-6α	H-4a, H-6α, H-12β				C-6a, C-7, C-10a, C-12, NMe
6a						129.9	s
7	6.61	d (8.0)	H-8	H-6α, H-8		122.4	d	C-6, C-9, C-10a
8	6.68	d (8.0)	H-7	H-7, OMe		111.8	d	C-6a, C-10
9						144.6	s
10						146.9	s
10a						128.9	s
10b						53.4	s
11α	4.05	dd (10.8, 4.0)	H-12α, H-12β	H-1, H-12α, H-12β, NMe		72.0	d	C-1, C-4a
12α	3.02	dd (14.0, 3.2)	H-11α, H-12β	H-11α, H-12β, NMe		62.1	t	C-6, C-10b, C-11
12β	3.17	dd (13.6, 10.8)	H-11α, H-12α	H-4a, H-6β, H-11α, H-12α				C-6, C-11, NMe
OMe	3.84	s (3H)		H-8		56.2	q	C-9
NMe	2.45	s (3H)		H-6α, H-11α, H-12α		43.2	q	C-6, C-12

3. Experimental

3.1. General

NMR spectra were recovered in a Varian Mercury 400 MHz instrument using CDCl₃ (CD₃OD for compound 5) as a solvent and TMS as the internal standard. Chemical shifts were reported in δ units (ppm) and coupling constants (J) in Hz. EIMS were obtained on a GC-MS Hewlett-Packard 6890+ MSD 5975 operating in EI mode at 70 Ev.

An HP-5 MS column (30 m × 0.25 mm × 0.25 µm) was used. The temperature program was: 100–180 °C at 15 °C min^–1, 1 min hold at 180 °C and 180–300 °C at 5 °C min^–1 and 10 min hold at 300 °C.Injector temperature was 280 °C. The flow rate of carrier gas (Helium) was 0.8 mL min^–1. Split ratio was 1:20. A QSTAR Elite hybrid Quadrupole-Time of Flight (QToF) mass spectrometer (Applied Biosystems, PE Sciex, Concord, ON, Canada) was used for HR-MS analysis. ToF MS data were recorded from m/z 70 to 700 amu with an accumulation time of 1 s and a pause between the mass range of 0.55 ms, operating in the positive mode. Reserpine (1 ρmol/µL) in product ion scan mode of m/z 609 was used for calibration of the mass spectrometer. Optical rotations were carried out on a Perkin-Elmer 241 polarimeter. A Jasco-J-810 Spectrophotometer was used to run CD spectra, all recorded in MeOH. UV spectra were obtained on a DINKO UV2310 instrument and IR spectra were recorded on a Nicolet Avatar 320 FT-IR spectrophotometer.

3.2. Plant Material

Hippeastrum papilio was collected during the flowering period (November, 2009) in the South of Brazil (Caxias do Sul-RS). A voucher specimen (ICN-149428) has been deposited in the Institute of Botany, Universidade do Rio Grande do Sul (UFRGS), Porto Alegre, and identified by Julie Dutilh PhD, University of Campinas.

3.3. Extraction and Isolation of Alkaloids

Fresh bulbs (2 Kg) were crushed and exhaustively extracted with EtOH (96% v/v) at room temperature for 48 h and the combined macerate was filtered and evaporated to dryness under reduced pressure. The bulb crude extract (50 g) was acidified to pH 2 with diluted H₂SO₄ and extracted with Et₂O (4 × 250 mL) to remove neutral material. The aqueous solution was basified with 25% ammonia up to pH 11 and extracted with n-hexane (8 × 250 mL) to give extract A (0.55 g). Another extraction using EtOAc (8 × 250 mL) gave extract B (1.2 g) and the last extraction using EtOAc-MeOH (3:1, 3 × 250 mL) gave extract C (3.4 g). Extract A yielded galanthamine (2) by crystallization from acetone. Extract B was subjected to a VLC column (3 × 6 cm) using silica gel (250 g – Kieselgel – mesh 0.15/0.30), eluting with n-hexane gradually enriched with EtOAc (0 → 100%) and then with MeOH (0→50%). Fractions of 100 mL were collected (190 in total) monitored by TLC (Dragendorff’s reagent, UV light λ 254 nm) and combined according to their TLC profiles, obtaining three fractions: 70–90 (fraction I), 100–124 (fraction II) and 125 – 145 (fraction III). From I, galanthamine (2, 150 mg, 0.0075% of fresh bulbs) was isolated again by crystallization from acetone. Fraction II (250 mg) was subjected to a VLC column (1,5 × 3,5 cm) using n-hexane gradually enriched with EtOAc (0–100%) and then with MeOH (0–50%), providing 100 fractions. After combining fractions 55–85, PTLC (20 cm × 20 cm × 0.25 mm, Silica gel F₂₅₄, EtOAc:CHCl₃:n-Hexane:MeOH = 4:2:2:1, v/v/v/v, in NH₃ atmosphere) was used to isolate haemanthamine (4, 80 mg) and 8-O-demethyl-maritidine (6, 3.5 mg). From III, using PTLC (20 cm × 20 cm × 0.25 mm, Silica gel F₂₅₄, EtOAc-CHCl₃-MeOH = 4:2:1, in NH₃ atmosphere) 11-hydroxyvittatine (5, 10mg) and 11β-hydroxy-galanthamine (1, 55 mg, 0.00275% of fresh bulbs) were isolated. Fresh leaves (approx. 1Kg) were also submitted to alkaloid extraction. Their alkaloid profile obtained by TLC and GC-MS was quite similar to that observed for bulbs, with additional traces of narwedine (3) and vittatine (7), which were identified by comparing their GC-EI-MS spectra and Kovats retention indices (RI) with our own library database. All known alkaloids isolated were identified by comparing their physical and spectroscopic data with those of alkaloids previously isolated and characterized by our group [4,15,16,17,18].

3.4. Microplate AChE Assay

The assay for measuring AChE activity was performed as described by López et al. [14]. Galanthamine hydrobromide was used as a positive control. The IC₅₀ of 11β-hydroxygalanthamine, galanthamine hydrobromide and galanthamine was measured in triplicate and the results are presented as a mean ± standard deviation using the software package Prism (Graph Pad Inc., San Diego, USA). Both isolated alkaloids and the positive control were evaluated at a concentration ranging from 10^–3 M to 10^–8 M.

11β-Hydroxygalanthamine (1). White needles. UV (MeOH) λ_max nm: 212.5, 287. [α]^D₂₄ = –20° (c 1.1, CHCl₃); CD [Θ]²⁰_λ: [Θ]₂₃₀ +586, [Θ]₂₄₇ –1950, [Θ]₂₉₁ +3782. IR (CHCl₃) ν_max cm^–1: 3360, 2925, 2854, 1730, 1624, 1590, 1508, 1440, 1280, 1096, 1044, 977, 756. ¹H-NMR, COSY, NOESY, HSQC, HMBC (400 MHz, CDCl₃) and ¹³C-NMR (100 MHz, CDCl₃) see Table 1. EI-MS 70eV (rel. int.): 303(M⁺, 32), 302(18), 286(11), 231(22), 230(100), 213(25), 181(12), 174(10), 97(96), 57(12). HR-QTOF-MS [M + H]⁺: 304.1550 (cald for C₁₇H₂₂NO₄, 304.1549).

4. Conclusions

Although the chemical synthesis of galanthamine has been achieved, its supply for clinical use still comes from natural sources. Hippeastrum papilio is able to produce great quantities of galanthamine but more studies on its genetic improvement, hybridization or in vitro culture are needed. Our search for new sources of galanthamine and acetylcholinesterase inhibitors has resulted in the isolation and identification of 11β-hydroxygalanthamine (1), a new acetylcholinesterase inhibitor alkaloid. The compound was structurally elucidated by 2D NMR, which allowed us to distinguish it from its epimer, habranthine. Galanthamine-type alkaloids are well-known for their inhibitory activity of the acetylcholinesterase enzyme. The action of galanthamine as an allosterically potentiating ligand in nicotinic acetylcholine receptors [5] and its ability to inhibit β-amyloid aggregation [19] could also play a role in successful AD therapy. The discovery of new galanthamine-type candidates is therefore of real interest for the future management of this disease.