Identification of Alkaloids from Hippeastrum aulicum (Ker Gawl.) Herb. (Amaryllidaceae) Using CGC-MS and Ambient Ionization Mass Spectrometry (PS-MS and LS-MS)

Abstract

Amaryllidaceae alkaloids are well-known isoquinolines which have demonstrated a wide range of biological activities such as antiviral, anticancer, acetylcholinesterase inhibition, antimalarial, among others. Mass spectrometry (MS) studies based on capillary gas chromatography (CGC), paper spray (PS), and leaf spray (LS) ionization were carried out for alkaloid investigation of the native Brazilian species Hippeastrum aulicum, along with nuclear magnetic resonance (NMR) techniques. Thirty-one alkaloids were identified including the new compound haemanthamine N-oxide. The results from PS- and LS-MS techniques were consistent with those observed in CGC-MS analysis. To the best of our knowledge, it is the first study combining NMR, CGC-MS and the ambient ionization-mass spectrometry (PS- and LS-MS) on Amaryllidaceae plants.

Keywords:
Amaryllidaceae; Hippeastrum aulicum; CGC-MS; PS-MS/LS-MS; isoquinoline alkaloids

Introduction

Amaryllidaceae is a well-known family of monocotyledons which are distributed widely over the temperate and warm regions of the world.¹ Amaryllidaceae plants are able to synthesize a specific group of isoquinoline alkaloids, which have demonstrated remarkable biological activities, such as antitumoral, antiviral, antiparasitic, acetylcholinesterase inhibitory, among others.¹ The outstanding feature of Amaryllidaceae plants is a consistent presence of a unique group of alkaloids, which have been isolated from all the genera of this family.¹ The current investigation on Amaryllidaceae alkaloids has focused on hyphenated techniques, which exploit the advantages of both chromatographic and spectral methods.^2,3 This strategy can be understood as a kind of dereplication process,⁴ which is very attractive in that it avoids labor-intensive chromatographic steps and analysis of worthless components.

Capillary gas chromatography-mass spectrometry (CGC-MS) has become the most successful technique for a dereplication approach to Amaryllidaceae alkaloids since these compounds have shown accurate detection under CGC-MS conditions. The construction of an in-home alkaloidal database based on electron impact-mass spectrometry fragmentation (EI-MS) and retention index allows a quick identification of known compounds.⁵ Although most Amaryllidaceae alkaloids are suitable for CGC-MS analysis, even as underivatized compounds, there are exceptions like alkaloids in the form of salts or N-oxide,⁶ which therefore require the support of other methodologies for correct identification and/or quantification.

An easier way of generating ions in mass spectrometry was introduced with the new ionization techniques, such as ambient mass spectrometry (or ambient MS).^7,8 These techniques constitute simplified and efficient alternatives for the detection and quantification of analytes directly from their natural environments (the "real world") or when placed on auxiliary surfaces.⁹

An important ambient ionization technique is paper spray (PS) ionization. PS analysis is performed by placing the sample in the middle of a triangular piece of paper, held by a metallic clip attached to a high voltage source that is positioned in front of the inlet orifice of the mass spectrometer (MS). The paper is moistened with a solvent and a high voltage is applied to the paper through the metal clip. Consequently, in the paper tip is formed a spray with charged droplets that go towards the entrance of the MS and are analyzed.^9,10

Recently, a variant technique of PS was introduced, the leaf spray (LS).The difference between paper spray and leaf spray is that the latter uses the sample itself (a plant tissue) for generating ions in gaseous phase. Ions can be generated in plant tissue without adding a solvent,¹¹ due to the natural juice present in fruit and vegetables. However, mass spectra with more intense signals and an improved signal/noise ratio can be obtained when a solvent is added.

The PS-MS and LS-MS have been applied in chemical identification of natural products present in coffee,¹² fruits,¹³ extra-virgin olive oils¹⁴ and herbal teas.¹⁵ Furthermore, LS-MS has also demonstrated excellent sensitivity for the direct identification of chemical species on the surface of leaves of plants such as Populus deltoids, Populus grandidentata,¹¹Hibiscus moscheutos, Hibiscus syriacus¹⁶ and Illicium anisatum.¹⁷

In the present work, indigenous Brazilian Hippeastrum aulicum was submitted to a classical phytochemical procedure assisted by CGC-MS, PS-MS and LS-MS. Thirty-one compounds were identified, including the new compound haemanthamine N-oxide, which was completely characterized by mono (1D) and bidimensional (2D) nuclear magnetic resonance (NMR) experiments. The results obtained via CGC-MS were compared with ambient mass spectrometry techniques (PS-MS and LS-MS).

Experimental

General experimental procedures

Column chromatography (CC) and vacuum liquid chromatography (VLC) were carried out using silica gel 60 (70-230 mesh, Merck) and silica gel 60 ACC (6-35 µm, Chromagel-SDS), respectively. For thin layer chromatography (TLC), commercial plates with silica gel F₂₅₄ as the stationary phase and dimensions of 20 cm × 20 cm × 0.20 mm and 20 cm × 20 cm × 0.25 mm were used for analytical and semi-preparative TLC (SPTLC), respectively. High performance liquid chromatography (HPLC) was performed on an Agilent G1311C-1260 quaternary pump coupled to a UV-Vis diode array (DAD), model G1315D-1260, using the semi-preparative column Zorbax RX-Sil (9.4 × 250 mm, 5 µm) and HPLC grade solvents. NMR spectra (nuclear magnetic resonance) were recovered on a Varian 400 MHz instrument using deuterated chloroform (CDCl₃) or deuterated methanol (CD₃OD) as solvents and tetramethylsilane (TMS) as the internal standard. The CGC-MS spectra (capillary gas chromatography-mass spectrometry) were obtained on a GC-17A Shimadzu CG-MS QP 5000 operating in the EI mode at 70 eV using a DB5 MS column (30 m × 0.25 mm × 0.25 µm). The temperature program was as follows: 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. The injector temperature was 280 °C. The flow rate of carrier gas (helium) was 0.8 mL min^-1, and the split ratio was 1:20. HRESIMS (high-resolution electrospray ionization mass spectrometry) was performed on 9.4 T FT-ICRMS (Solarix) by direct injection of the compound dissolved in methanol (MeOH). A Jasco-J-810 Spectrophotometer (Easton, MD, USA) was used to run CD (circular dichroism) spectra, all recorded in MeOH. Infrared (IR) spectrum was recorded on a PerkinElmer Spectrum 400 FT-IR/FT-NIR Spectrometer. UV (ultraviolet) spectrum was obtained on a UV-PerkinElmer, Lambda 45, UV-Vis.

Acetone (Me₂CO), ammonia (NH₃), ammonium hydroxide (NH₄OH), 1-butanol (n-BuOH), dichloromethane (CH₂Cl₂), n-hexane (n-Hex), ethyl acetate (EtOAc) and sulphuric acid (H₂SO₄) used for the extraction and isolation procedures were of analytical grade.

Plant material

Approximately 1.7 kg of fresh bulbs and 1.0 kg of fresh leaves of Hippeastrum aulicum (Ker Gawl.) Herb. were collected in Biritiba-Mirim City of São Paulo State, in September 2013. A voucher specimen was deposited in the Herbarium UEC (Campinas-SP, Brazil), under the reference number 114. The species was identified by Dr Renata S. de Oliveira. A new specimen was collected at the same location in February 2016 and again identified as H. aulicum (Ker Gawl.) Herb. by Dr Renata S. de Oliveira. This specimen was used for ambient MS analysis.

Extract procedure

Fresh bulbs and leaves from H. aulicum were crushed and extracted with MeOH and the mixture was immediately filtered and the solvent evaporated under reduced pressure. The plant material was then twice extracted with MeOH (48 hours each), filtered and the solvent evaporated under reduced pressure. The remaining crude extract was finally combined.

Extraction and isolation procedures

Leaves

The leaf crude extract was acidified with H₂SO₄ (2%) up to pH 2 and extracted with n-Hex (9 × 200 mL) to remove neutral material. The aqueous phase was basified with NH₄OH (25%) up to pH 10 and extracted with n-Hex (5 × 200 mL) yielding the extract IA (166.5 mg), followed by the extraction with EtOAc (15 × 200 mL) affording extract IIA (897.2 mg), and finally extracted with the mixture of EtOAc:MeOH (3:1, 3 × 200 mL) providing the extract IIIA (1.56 g).

The extract IA was chromatographed by CC, starting with a mixture of EtOAc:n-Hex (9:1), gradually increasing the polarity up to 100% EtOAc and finally increasing the MeOH percentage in the mixture up to a ratio of EtOAc:MeOH (1:1). 320 fractions (5 mL each) were collected and after analytical TLC were grouped by similarity in six fractions. Fraction 5 was resuspended in MeOH, and haemanthamine (16) (5.0 mg) precipitated spontaneously. Fraction 6 was purified by SPTLC (n-Hex:EtOAc:Me₂CO:MeOH:isopropanol, 5:2:2:1:2, in NH₃ atmosphere) and allowed the isolation of albomaculine (29) (5.9 mg).

A CC column (EtOAc:MeOH, 49:1) was performed to purify the components of the extract IIA, in which 800 fractions (4 mL each) were collected and combined according to their TLC profile, which afforded 11 fractions. Fraction 9 (650.5 mg) was resuspended in MeOH and haemanthamine (16) (228.2 mg) precipitated spontaneously. The supernatant was subjected to CC, starting with EtOAc:MeOH (23:2) and increasing solvent polarity with MeOH up to 1:1, which allowed the collection of 342 fractions (4 mL each). Fractions were compared by analytical TLC (plates were revealed with Dragendorff's reagent and UV light at 254 nm) and combined according to their similarities, obtaining 10 subfractions. From subfraction 9.7 (320.5 mg, viscous material), 66.0 mg of haemanthamine (16) precipitated spontaneously. The supernatant from subfraction 9.7 was purified by CC starting with EtOAc:MeOH (9:1), gradually increasing the MeOH percentage up to EtOAc:MeOH (1:1) and affording 634 fractions (3 mL each). After analytical TLC analysis, fractions were again combined by similarity, yielding 12 subfractions. Sample 9.8.7 was purified by SPTLC (EtOAc:CH₂Cl₂:Me₂CO:MeOH:n-Hex, 2:2:2:1:1, along with drops of NH₄OH, in NH₃ atmosphere) allowing the purification of the alkaloids haemanthidine (22) and 6-epihaemanthidine (23) (16.5 mg).

Fractions 10 (33.9 mg) and 11 (34.5 mg) were purified by SPTLC (n-Hex:EtOAc:Me₂CO:MeOH:n-BuOH, 4:3:3:2:1, in NH₃ atmosphere). Haemanthamine N-oxide (1) (10.0 mg) and 7-methoxy-O-methyllycorenine (12) (5.7 mg) were obtained from fraction 10 and compound 12 (6.7 mg) was again obtained from fraction 11.

The extract IIIA showed a negligible alkaloid content by CGC-MS and analytical TLC analysis.

Bulbs

The bulb extract was submitted to the same acid-base extraction as outlined previously. The n-Hex extract (IB) afforded 506.7 mg, while the EtOAc extract (IIB) yielded 2.65 g. Finally, 2.06 g was provided by the EtOAc:MeOH (3:1) extract (IIIB).

The extract IB was subjected to CC eluting, firstly with a mixture of EtOAc:MeOH:CH₂Cl₂ (3:1:1), along with drops of the NH₄OH. The system polarity was gradually increased, enriching the mixture with MeOH and decreasing the amount of EtOAc to 50% of MeOH in the mixture. Approximately 770 fractions of 5 mL each were collected. These fractions were analyzed by TLC and then combined according to their alkaloidal profile under UV λ 254 light and Dragendorff's reactive spots, affording 18 fractions. Fraction 9 (60.0 mg) was purified by SPTLC (CH₂Cl₂:EtOAc:Me₂CO:n-Hex:MeOH, 2:2:1.5:3:1, along with drops of NH₄OH, in NH₃ atmosphere) and alkaloids haemanthamine (16) (6.5 mg), albomacuine (29) (3.5 mg) and 7-methoxy-O-methyllycorenine (12) (8.0 mg) were isolated. The alkaloid aulicine (15) (48.0 mg) precipitated from fraction 11.

After resuspension of the extract IIB in MeOH, lycorine (25) (160.7 mg) precipitated spontaneously. The supernatant (1.94 g) was then subjected to a VLC column (4.4 × 7.0 cm), starting with 100% of n-Hex, gradually enriching with EtOAc (0-100%) and finally with MeOH (0-50%). A total of 216 fractions (50 mL each) was obtained and combined according to their alkaloid profile by TLC (UV 256 and Dragendorff's reactive spots), affording 12 fractions. The alkaloid trisphaeridine (3) (1.5 mg) was isolated by HPLC from fraction 4 using a normal phase column (9.4 × 250 mm) and EtOAc:n-Hex (4:1) mixture as a mobile phase under a flow rate of 0.75 mL min^-1. In fraction 12 (1.17 g) the alkaloid haemanthamine (16) (251.6 mg) precipitated after resuspension in MeOH. The supernatant was purified by CC and the system elution started with a mixture of EtOAc:CH₂Cl₂:Me₂CO:MeOH:n-Hex (2:2:1:1:1), the system polarity being increased by adding MeOH (up to the ratio 2:2:1:5:1). 944 fractions of 4 mL each were obtained. The fractions were combined according to their alkaloid profile by TLC using UV (λ 256) and Dragendorff's reagent stain, resulting in 11 subfractions. SPTLC (EtOAc:CH₂Cl₂:Me₂CO:MeOH:n-Hex, 2:2:1:1:1, along with drops of the NH₄OH, in NH₃ atmosphere) was performed to purify subfraction 13.6, and the alkaloid galanthine (21) (3.6 mg) was isolated. Subfraction 13.7 was identified as the alkaloid haemanthamine (16) (49.0 mg). After resuspension of subfraction 13.8 in MeOH, haemanthamine (16) (253.0 mg) precipitated spontaneously. Subfraction 13.9 was purified by SPTLC (EtOAc:CH₂Cl₂:Me₂CO:MeOH:n-Hex, 2:2:1:1:1, along with drops of the NH₄OH, in NH₃ atmosphere) resulting in the isolation of the isomers haemanthidine (22) and 6-epihaemanthidine (23) (3.5 mg) and alkaloid norpluviine (14) (1.5 mg).

Like extract IIIA, extract IIIB showed a negligible alkaloid content by CGC-MS and analytical TLC analysis.

Haemanthamine N-oxide (1)

Amorphous solid; CD [Θ]_λ ²⁰: [Θ]₂₄₅ - 5224, [Θ]₂₉₄ + 2788; UV (MeOH) λ_max / nm (ε) 237 (1692), 292 (1884); IR (CHCl₃) ν_max / cm^-1 3361, 2926, 1640, 1487, 1247, 1037, 750; ¹H NMR (400 MHz, CDCl₃) and ¹³C NMR (100 MHz, CDCl₃) see Table 1; HRESIMS of [M + H]⁺ m/z 318.13362 (calculated for C₁₇H₂₀NO₅: 318.13360).

Table 1
¹H NMR, COSY, NOESY, HSQC and HMBC data of haemanthamine N-oxide (1) (400 MHz, CD₃OD)

	δH ( J in Hz)	COSY	NOESY	HSQC/13C	HMBC
1	6.36 br s		H-10	126.8 d	C-3, C-4a, C-10b
2	6.36 br s	H-3, H-4β	H-3	131.5 d	C-3, C-10b
3	4.02 br s	H-2, H-4α, H-4β	H-2, 3-OMe, H-4α, H-4β	72.9 d	C-1, C-2
4α	2.36 ddd (13.6, 13.2, 4.4)	H-3, H-4a, H-4β	H-3, H-12exo, H-4β	24.7 t	C-4a
4β	2.75 ddd (13.6, 4.0, 1.6)	H-2, H-3, H-4a, H-4α	H-3, H-4a, H-4α		C-2, C-3, C-4a, C-10b
4a	3.74 dd (13.2, 3.6)	H-4α, H-4β	H-4β, H-6β	75.2 d	C-11
6α	4.68 d (15.4)	H-7	H-7, H-12endo	75.4 t	C-4a, C-6a, C-7, C-10a
6β	4.75 d (15.4)	H-7, H-12exo	H-4a, H-7		C-6a, C-7, C-10a, C-12
6a				122.5 s
7	6.68 s	H-6α, H-6β	H-6α, H-6β	107.3 d	C-6, C-9, C-10a
8				148.9 s
9				149.6 s
10	6.98 s		H-1	104.5 d	C-6a, C-8, C-10b
10a				133.5 s
10b				53.7 s
11endo	3.94 br dd (7.2, 3.6)	H-12endo, H-12exo	H-12endo	77.3 d
12endo	4.37 dd (13.6, 7.2)	H-11endo, H-12exo	H-6α, H-11 endo, H-12exo	76.9 t	C-4a, C-6, C-10b
12exo	3.71-3.66 m	H-6β, H- 11 endo, H-12endo	H-4α, H-12endo
3-OMe	3.39 s		H-3	57.0 q	C-3
OCH2O	5.97 s			103.0 t	C-8, C-9

• COSY: correlation spectroscopy; NOESY: nuclear Overhauser effect spectroscopy; HSQC: heteronuclear single quantum coherence; HMBC: heteronuclear multiple bond correlation.

Identification of alkaloids by CGC-MS

The alkaloids were identified by comparing their CGC-MS spectra and Kovats retention indices (RI) with our library database. This library has been regularly updated with alkaloids isolated and unequivocally identified via physical and spectroscopic methods.¹⁸ NMR data for the known alkaloids described herein closely matched those reported elsewhere.^18-26 Mass spectra were deconvoluted using AMDIS 2.64 software (NIST) (WA, USA) and RIs recorded using a standard n-hydrocarbon calibration mixture (C9-C36). The proportion of individual components in the alkaloid fractions are expressed as a percentage of total alkaloid content. CGC-MS peak areas are dependent on the concentration of the injected alkaloid as well as the intensity of its mass spectral fragmentation. Although the data given in Table 2 are not representative of a validated alkaloid quantification method, they can be used for relative comparison purposes.

Table 2
CGC-MS data for H. aulicum. Values are expressed as a relative percentage of total ion current (TIC)

Alkaloid	RI	Bulbsa / %		Leavesb / %		M+	MS
		IB	IIB	IA	IIA
Ismine (2)	2280	-	trd	-	trd	257(35)	238(100), 211(6), 196(8), 168(6), 154(3), 106(4), 77(3)
Trisphaeridine (3)	2282	0.13	0.47	0.10	0.54	223(100)	222(38), 167(8), 165(9), 164(14), 138(20), 137(9), 111(13)
Galanthamine (4)	2395	6.43	6.11	0.34	trd	287(83)	286(100), 270(13), 244(24), 230(12), 216(33), 174(27), 115(12)
Lycoramine (5)	2420	trd	trd	-	-	289(62)	288(100), 232(8), 202(14), 187(14), 159(9), 115(19)
Buphanisine (6)	2424	trd	-	-	-	285(100)	270 (33), 254(34), 215 (85), 201(24), 172(19), 157(21), 115(33)
Vitattine (7)	2472	-	0.55	-	-	271(100)	228(25), 199(95), 187(85), 173(28), 128(32), 115(33), 56(22)
Narwedine (8)	2483	0.22	trd	trd	-	285(84)	284(100), 242(18), 216(20), 199(18), 174(31), 128(16), 115(16)
O-Methyllycorenine (9)	2492	0.65	-	-	-	331(< 1)	300(3), 191(8), 147(1), 110(8), 109(100), 94(3), 77(1)
Nerinine (10)	2509	-	0.86	-	-	347(< 1)	222(1), 207(2), 179(1), 164(1), 110(8), 109(100), 108(18), 94(2)
8-O-Demethylmaritidine (11)	2510	-	1.27	-	0.83	273(100)	256(22), 230(20), 201(83), 189(42), 174(22), 128(23), 115(24)
7-Methoxy-O-methyllycorenine (12)	2538	17.71	-	-	trd	361(< 1)	330(8), 221(10), 191(2), 110(8), 109(100), 108(15), 94(2), 83(2)
11-Oxohaemanthamine (13)	2585	-	trd	-	trd	299(< 1)	271(100), 270(37), 240(10), 238(10), 211(23), 181(77), 152(20)
Norpluviine (14)	2596	-	trd	-	-	273(68)	272(41), 254(45), 229(45), 228(100), 214(13), 186(10), 147(8), 91(8), 77(12)
Aulicine (15)	2607	44.24	12.06	60.62	3.17	319(100)	304(19), 288(37), 246(18), 233(73), 218(19), 206(26), 163(7)
Haemanthamine (16)	2641	14.66	71.09	20.60	92.0	301(14)	272(100), 257(10), 240(16), 181(21), 214(12), 211(14), 128(8)
Crinamine (17)	2648	-	-	0.27	-	301(< 1)	272(100), 242(10), 211(17), 181(23), 153(14), 128(18), 115(16), 77(6)
Tazettine (18)/Pretazettine (19)c	2653	0.44	1.57	trd	0.61	331(31)	316(15), 298(23), 247(100), 230(12), 201(15), 181(11), 152(7)
Hamayne (20)	2699	-	-	-	0.19	287(3)	258(100), 242(6), 211(12), 186(17), 181(11), 153(10), 128(19)
Galanthine (21)	2709	-	0.41	-	0.33	317(22)	316(15), 298(10), 268(18), 243(96), 242(100), 228(8)
Haemanthidine (22)/6-Epihaemanthidine (23)	2716	-	trd	-	trd	317(59)	284(52), 233(48), 211(45), 201(80), 199(70), 181(69), 173(71), 115(10), 56(71)
11-Hydroxyvittatine (24)	2728	-	0.60	-	0.40	287(5)	258(100), 211(15), 186(20), 181(23), 153(13), 128(24), 115(23)
Lycorine (25)	2746	-	0.71	-	trd	287(31)	286(19), 268(24), 250(15), 227(79), 226(100), 211(7), 147(15)
Incartine (26)	2756	-	1.24	trd	trd	333(42)	332(100), 315(25), 259(73), 258(97), 244(17), 242(6), 214(9), 172(45)
Homolycorine (27)	2765	6.52	1.68	trd	1.82	315(< 1)	206(< 1), 178(2), 109(100), 150(1), 108(22), 94(3), 82(3)
3-Epimacronine (28)	2811	trd	0.25	-	trd	329(27)	314(23), 245(100), 225(14), 201(83), 139(16), 70(18)
Albomaculine (29)	2815	8.93	1.08	3.10	-	345(< 1)	221(1), 193(1), 165(1), 110(10), 109(100), 108(25), 94(2), 82(3)
8-O-Demethylhomolycorine (30)	2841	-	-	14.94	-	301(< 1)	195(0.5), 164(2), 109(100), 108(25), 94(3), 82(3)
2α-Methoxyhomolycorine (31)e	2870	-	-	trd	-	345(< 1)	206(< 1), 178(2), 150(1), 139(100), 124(64), 96(5), 94(5), 81(3)

• ^a
  Alkaloid percentage in the total mixture of alkaloids from n-Hex and EtOAc fractions of bulbs (IB and IIB, respectively);
• ^b
  alkaloid percentage in the total mixture of alkaloids from n-Hex and EtOAc fractions of leaves (IA and IIA, respectively);
• ^c
  tazettine detection by CGC-MS mean identification of both alkaloids tazettine (18) and pretazettine (19);³¹
• ^d
  traces < 0.05 of TIC. The alkaloid Haemanthamine N-oxide (1) is not suitable to CGC-MS detection;
• ^e
  alkaloids identified using NIST 05 database; recursive procedure, HR-MS and literature data. RI: Kovats retention indices.

PS-MS and LS-MS

For LS-MS analysis, freshly collected leaves and bulbs of H. aulicum were cut into a triangle (base and height of 1 cm each) and held by a metal clip at a distance of 5-7 mm from the mass spectrometer inlet (Figure 1a).⁹ Approximately 10 µL of MeOH (HPLC grade, JTBaker) and a high voltage (3 kV) supplied by the mass spectrometer were applied to the leaf or bulb to generate the LS mass spectra.

Figure 1
Scheme of (a) LS(+)-FT-ICR MS and (b) PS(+)-FT-ICR MS analyzes.

For PS-MS analysis, extracts of bulbs and leaves were dissolved in MeOH at 2 mg mL^-1. Then, 10 µL of solution was applied on the surface of a triangular paper (Whatman Grade 1, GE Healthcare)^27,28 with base and height of 1 cm each. The triangular paper was fixed with a metal clip, connected to 0.5 mm wire linked to the mass spectrometer (Figure 1b). Then, 20 µL of MeOH was applied with a high voltage (3 kV) to the triangular paper to generate the PS mass spectra.

LS and PS-MS experiments were performed in positive ion mode (LS(+) and PS(+)) using a Fourier transform ion cyclotron resonance mass spectrometer (FT-ICR MS, model 9.4 T Solarix, Bruker Daltonics Bremen).^27,29 Ion time accumulation was 0.010 s. LS(+) and PS(+)-FT-ICR mass spectra were acquired by accumulating 32 scans of time-domain transient signals in 16 mega-point time-domain data sets. All mass spectra were externally calibrated using NaTFA (m/z from 200 to 1200). A resolving power, m /Δm _50% = 78000 (in which Δm _50% is the full peak width at the half-maximum peak height of m/z 300) and a mass accuracy of < 2 ppm provided the unambiguous molecular formula assignments for singly charged molecular ions. The proposed structures for each formula were assigned using the chemspider (www.chemspider.com) database. The degree of unsaturation for each molecule can be deduced directly from its double bond equivalent (DBE) value according to equation DBE = c - h / 2 + n / 2 + 1, where c, h, and n are the numbers of carbon, hydrogen, and nitrogen atoms, respectively, in the molecular formula.²⁹

Results and Discussion

Alkaloid comparison

The phytochemical procedure assisted by MS and NMR approaches identified thirty-one compounds in H. aulicum (Figure 2 and Table 2). Thirteen alkaloids are reported here for the first time in H. aulicum, although some chemical similarities with previous studies have also been found.^18,30 The H. aulicum studied here, from Biritiba-Mirim City (São Paulo, Brazil), displayed aulicine (15), haemanthamine (16), and lycorine (25) as the main compounds, as does H. aulicum from Cunha City (São Paulo, Brazil).¹⁸ Both Brazilian cities are relatively close, which may explain the presence of the same major components. Lycorine (25) was found to be one of the main alkaloids (see Experimental section) even though its low solubility in MeOH³¹ covered the correct relative quantification by CGC-MS (Table 2). A former H. aulicum investigation³⁰ also revealed alkaloids such as norpluviine (14) and galanthine (21), which were observed in the present work. No information about the collection of the plant species in this previous study is available. In contrast, while ambelline, anhydrolycorine, chlidanthine, montanine, narcissidine and pseudolycorine have been previously reported,^18,30 they were not found in this work. Notably, the identification of the alkaloids incartine (26) and buphanisine (6), which are uncommon in the Hippeastrum genus, is reported here for the first time in H. aulicum.³²

Figure 2
Alkaloids identified in Hippeastrum aulicum.

CGC-MS dereplication

The CGC-MS results are shown in Table 2. The specific EI-MS fragmentation mechanisms for the distinct skeleton types together with retention indices are the key for alkaloid identification in Amaryllidaceae research. Concerning the skeleton types found in H. aulicum, the EI mass fragmentation of homolycorine-type alkaloid Δ^3,4-derivatives features a dominant retro-Diels Alder process and cleavage of ring C, yielding a very abundant ion peak characterized by the pyrrolidine ring (m/z 109).³³ Consequently, the alkaloids 9, 10, 12, 27, 29 and 30 show the base peak at m/z 109 (100%), with all remaining ion peaks displaying less than 10% of abundance. Otherwise, alkaloid 31 possesses a methoxyl group at C-2 and so displays the base peak at m/z 139, which is in agreement with the pyrrolidine residue along with the methoxyl substituent at C-2.³³ Lycorine-type compounds, which are biogenetically related to the homolycorine skeleton, also suffer a retro-Diels Alder process followed by the loss of C-1 and C-2, along with their substituents, yielding the base peak at m/z 228, 242 and 226 for compounds 14, 21 and 25, respectively. Alkaloid 26, which lacks the 3,4-unsaturation, possesses the base peak at m/z 332 (M-1) instead of the typical base peak after the retro-Diels Alder process.³⁴

The EI-MS fragmentation of the tazettine-type skeleton, represented by compounds 18, 19 and 28, is strongly supported by the stereochemistry of the substituent at C-3. The b-orientation of the C-3 substituent induces a retro-Diels Alder process at ring C followed by the loss of the neutral fragment [C₅H₈O], which yields the base peaks at m/z 247 and 245 for compounds 18 and 28, respectively (Table 2). Compound 18 is an artefact of 19 under CGC-MS.³¹ Considering the high temperature of the CGC-MS method, the haemanthamine-type skeleton also suffers a thermal decomposition, particularly those compounds with a substitution at C-11, as in the case of 13, 16, 17, 20, 22, 23 and 24.^6,35 Conversely, the EI-MS fragmentation of galanthamine-type compounds under CGC and direct insertion probe (DIP) conditions are very similar and feature abundant [M]⁺ and [M - H]⁺ ion peaks, similarly to lycorine-type alkaloids.³⁶ The galanthamine-type compounds also show typical fragmentation, with domination of one or another pathway depending on the substituents of the skeleton. In summary, the 4,4a-unsaturation induces the loss of the substituent at C-3 followed by the elimination of the nitrogen [C₃H₇N], as in galanthamine (4) and narwedine (8). In the EI-MS spectrum of lycoramine (5), a dihydro derivative, this induction is weaker and the remaining ion peaks are considerably less abundant than those observed for galanthamine (4) and narwedine (8) (Table 2).

Phytochemical procedure and NMR data of haemanthamine N-oxide (1)

In the course of the phytochemical procedure, the known Amaryllidaceae alkaloids haemanthamine (16),^19,23 albomaculine (29),¹⁸ haemanthidine (22), 6-epihaemanthidine (23),¹⁹ 7-methoxy-O-methyllycorenine (12),¹⁸ aulicine (15),¹⁸ lycorine (25),^22,23 trisphaeridine (3),^21,24 galanthine (21),²⁰ and norpluviine (14),²⁵ were isolated and identified by comparison of their spectroscopic data with those previously reported. The phytochemical procedure was assisted by CGC-MS and the new compound haemanthamine N-oxide (1) was purified from a natural source for the first time. Compound 1 showed an ¹H NMR spectrum very similar to that of haemanthamine. The characteristic ¹H NMR spectrum included: (i) two para-oriented aryl singlets at δ 6.98 and 6.68, the most deshielded of which was assigned to H-10 due to its NOESY (nuclear Overhauser effect spectroscopy) correlation with H-1; (ii) H-1 and H-2 were assigned at δ 6.36 (2H), confirmed by COSY (correlation spectroscopy) correlation with H-3 (δ 4.02) and two HSQC (heteronuclear single quantum coherence) correlations at δ 126.8 and 131.5, respectively; (iii) a typical resonance at δ 5.97 corresponding to the methylenedioxy group; (iv) the aliphatic methoxyl group confirmed at position C-3 due to the HMBC (heteronuclear multiple bond correlation) correlation between 3-OMe and C-3; (v) the magnitude of the coupling constant between H-4α and H-3 (J = 4.4 Hz) was indicative of a trans relationship between 3-OMe and the 5,10b-ethano bridge. The remaining ¹H NMR signals were consistent with the structure of haemanthamine (16), with the exception of positions H-4a, H-6 and H-12. The H-4a resonance was 0.48 ppm more deshielded than its homolog in haemanthamine. Both H-6 were assigned at δ 4.68 and 4.75 ppm, 0.94 and 0.49 ppm more deshielded than the H-6α and H-6β in haemanthamine, respectively. Finally, the H-12endo and H-12exo were 0.94 and 0.50 ppm more deshielded than their homologs in haemanthamine. This kind of deshielding effect was congruent with the alkaloid haemanthamine in salt or N-oxide form. HRESIMS analysis was carried out to confirm an additional oxygen atom in the structure. Compound 1 exhibited a parent [M + H]⁺ ion at m/z 318.13362 in its HRESIMS spectrum, suggesting the molecular formula C₁₇H₂₀NO₅ (calcd. 318.13360), and confirming 1 as haemanthamine N-oxide. Haemanthamine N-oxide (1) was obtained here for the first time from a natural source, even though it has already been obtained by synthesis.³⁷ The complete 1D and 2D NMR data are shown in Table 1.

The confirmation of the absolute stereochemistry of 1 was achieved through circular dichroism (CD). Negative and positive Cotton effects were observed at 245 and 294 nm, whose absorptions were in agreement with the haemanthamine-type skeleton.³⁸

LS(+)-MS and PS(+)-MS

Figure 3 displays the LS(+) mass spectra for bulb and leaf analyses of H. aulicum, which had a similar chemical profile. Signals varying from m/z 280 to 400 were detected as protonated molecules, [M + H]⁺. A similar m/z distribution has been observed for Hibiscus species.¹⁶ Table 3 shows measured m/ z values, mass error (ppm), DBE and molecular formula of the main compounds detected by the LS(+)-MS technique, where 10 and 8 species were identified in bulb and leaf, respectively. The ions of m/z 302, 318, 320, 332 and 348 are the most abundant species found, having double bond equivalents (DBEs) of 7-9. These species correspond to alkaloids, presenting odd molecular weight values from molecular formula of neutral species (M). As a consequence, an odd number of nitrogen is detected in their chemical structure, N₁O_x class, where x = 4-6 (Table 3). The species of m/z 302.1387 and 320.1858 ([C₁₇H₁₉NO₄ + H]⁺ and [C₁₈H₂₅NO₄ + H]⁺ ions) correspond to haemanthamine (16) and/or its isomers and aulicine (15), respectively (Figure 1). The high relative intensity detected for these species is in good agreement with the CGC-MS data (Table 2). Compounds with m/z 318.1337, 332.1492 and 348.1809 ([C₁₇H₁₉NO₅ + H]⁺, [C₁₈H₂₁NO₅ + H]⁺ and [C₁₉H₂₅NO₅ + H]⁺ ions, respectively) correspond to haemanthamine N-oxide (1) and its isomer, tazettine/petrazettine (18-19), and nerinine (10), respectively. The first compound is an isobar of galanthine (21), M = C₁₈H₂₃NO₄, which was only detected by the CGC-MS technique (Table 2). The ion of m/z 288.1233, detected only in bulbs, corresponds to the [C₁₆H₁₇NO₄ + H]⁺ ion, that is, lycorine (25), 11-hydroxyvittatine (24) or hamayne (20). The galanthamine (4) isobar, M = C₁₇H₂₁NO₃ and M_w 287 Da, was not identified, as indicated in Table 2. Another alkaloid was detected from LS(+)-MS as a [C₁₈H₂₅NO₅ + H]⁺ ion of m/z 336 (Table 3), although no chemical structure has been proposed so far. The only species not classified as an alkaloid was detected by LS(+)-MS as a sugar: an adduct sucrose, [C₁₂H₂₂O₁₁ + K]⁺ ion of m/z 381.0801.

Figure 3
LS(+)-FT-ICR mass spectra for bulb and leaf analysis of H. aulicum.

Table 3
Measured m/z values, mass error (ppm), DBE and molecular formula of main compounds detected from LS(+)-MS and PS(+)-MS data for H. aulicum

Proposed alkaloid	LS-MSa		PS-MSb				m/zmeasured	Error / ppm	DBEe	[M + H]+
	Bulb	Leaf	Bulbsc		Leavesd
			IB	IIB	IA	IIA
Hamayne (20) or 11-hydroxyvittatine (24) or lycorine (25)	X	-	-	X	-	-	288.12288	0.55	9	[C16H17NO4 + H]+
Haemanthamine (16) or crinamine (17) or 8-O-demethylhomolycorine (30)	X	X	X	X	X	X	302.13868	0.02	9	[C17H19NO4 + H]+
Homolycorine (27)	X	X	-	-	-	-	316.15443	0.30	9	[C18H21NO4 + H]+
Haemanthidine (22) or 6-epihaemanthidine (23) or haemanthamine N-oxide (1)	X	X	X	X	-	X	318.13369	0.29	9	[C17H19NO5 + H]+
	-	X	-	-	-	-	356.08986	1.06		[C17H19NO5 + K]+
Aulicine (15)	X	X	X	X	X	X	320.18577	0.42	7	[C18H25NO4 + H]+
Tazettine (18) or pretazettine (19)	X	X	X	X	-	X	332.14924	0.03	9	[C18H21NO5 + H]+
-	X	-	X	X	X	X	336.18058	0.08	7	[C18H25NO5 + H]+
2α-Methoxyhomolycorine (31)	X	X	X	-	-	X	346.16562	0.72	9	[C19H23NO5 + H]+
Nerinine (10)	X	X	X	X	X	X	348.18086	0.91	8	[C19H25NO5 + H]+
7-Methoxy-O-methyllycorenine (12)	-	-	X	X	X	X	362.19673	1.46	8	[C20H27NO5 + H]+
-	-	-	-	X	X	-	376.17629	2.19	9	[C20H25NO6 + H]+
Sucrose or isomers	X	X	-	-	-	-	381.08008	1.86	2	[C12H22O11 + K]+

• ^a
  Leaf spray mass spectrometry;
• ^b
  paper spray mass spectrometry;
• ^c
  alkaloid in the total mixture from n-Hex and EtOAc fractions of bulbs (IB and IIA, respectively);
• ^d
  alkaloid in the total mixture from n-Hex and EtOAc fractions of leaves (IA and IIA, respectively);
• ^e
  double bond equivalent.

Figure 4 shows PS(+)-FT-ICR mass spectra from n-Hex and EtOAc extracts of bulbs and leaves for H. aulicum. The chemical profiles for extracts of bulbs and leaves are quite similar. However, the PS(+) technique promotes a selective ionization for compounds of m/z ≥ 332 for bulb extracts (n-Hex and EtOAc), while in leaf extracts, two alkaloids are selectively identified: ions of m/z 362 ([C₂₀H₂₇NO₅ + H]⁺), and 376 ([C₂₀H₂₅NO₆ + H]⁺). Among them, the ion of m/z 362 (7-methoxy-O-methyllycorenine (12)) is also listed in Table 2. Figures 5a and 5b compare the relative intensity of most ions detected (m/z 288, 302, 316, 318, 320, 332, 336, 346, 348, 356, 374, 376, 362 and 381) by both LS(+)-MS (Figure 5a) and PS(+)-MS (Figure 5b). In general, a good agreement is observed between the ambient MS techniques. LS(+)-MS has proved to be a promising approach for the identification of alkaloids from Hippeastrum aulicum (Ker Gawl.) Herb. and should be explored in future work with other species, since it is a fast and easy analysis requiring no prior step of sample preparation.

Figure 4
PS(+)-FT-ICR mass spectra from n-Hex and EtOAc extracts of bulbs and leaves for H. aulicum.

Figure 5
Comparison of relative intensity of main ions detected from LS(+) and PS(+)-FT-ICR MS of bulbs and leaves for H. aulicum.

Conclusions

The phytochemical investigation of Hippeastrum aulicum resulted in the identification of thirty-one alkaloids and the CGC-MS dereplication proved to be very useful for the fast identification of a great number of compounds from an alkaloid-rich H. aulicum extract. The phytochemical fractionation assisted by CGC-MS analysis allowed the isolation of the new compound haemanthamine N-oxide (1), reported for the first time from a natural source.

Paper and leaf spray ionization mass spectrometry constitute a new family of ionization techniques able to identify alkaloids directly from their natural environments, i.e., in the "real world" of analytes (as demonstrated by LS-MS) or when placed on auxiliary surfaces such as PS-MS. A total of 13 species were identified with m/z ranging from 200 to 400, DBEs of 7-10 and exact mass lower than 2 ppm. Regarding the chemical structure of alkaloids, their carbon number varied from C₁₆ to C₂₀, containing NO_x as the heteroatom class, where x = 4-6. Among the main species detected were compounds with m/z 302, 318, 320, 332 and 348, which correspond to haemanthamine (16) and/or its isomers (17 and 30), haemanthamine N-oxide (1) and its isomers (22 and 23), aulicine (15), tazettine/pretazettine (18 and 19), and nerinine (10), respectively. Similar LS(+)-MS and PS(+)-MS results were obtained from leaf and bulb surfaces. Taken together, the results obtained from ambient ionization MS demonstrated a notable agreement with the CGC-MS analysis.

Supplementary Information

Spectra and NMR data of all isolated alkaloids are available free of charge at http://jbcs.sbq.org.br.

https://minio.scielo.br/documentstore/1678-4790/nFthcJm6SWbJmKTLbDwGcxt/477443f6aa567eb0b5159f58cd9f868e47f5eb75.pdf

• FAPESP has sponsored the publication of this article.

Acknowledgments

The authors thank R. P. Camargo for collecting the plant material. The authors also thank CAPES-PVE (No. 88881.030427/2013-01), CNPq (445987/2014-6), CAPES (23038.007083/2014-40), FAPES (65921380/2013) and NCQP-UFES.

References

• ¹
  Bastida, J.; Berkov, S.; Torras-Claveria, L.; Pigni, N. B.; de Andrade, J. P.; Martínez, V.; Codina, C.; Viladomat, F. In Recent Advances in Pharmaceutical Sciences; Muñoz-Torrero, D., ed.; Transworld Research Network: Kerala, 2011.
• ²
  Cortes, N.; Alvarez, R.; Osorio, E. H.; Alzate, F.; Berkov, S.; Osorio, E.; J. Pharm. Biomed. Anal. 2015, 102, 222.
• ³
  Shawky, E.; Abou-Donia, A. H.; Darwish, F. A.; Toaima, S. M.; Takla, S. S.; Pigni, N. B.; Bastida, J.; Chem. Biodiversity 2015, 12, 1184.
• ⁴
  Hubert, J.; Nuzillard, J.-M.; Renault, J.-H.; Phytochem. Rev. 2015, DOI 10.1007/s11101-015-9448-7.
  » https://doi.org/10.1007/s11101-015-9448-7
• ⁵
  Berkov, S.; Bastida, J.; Sidjimova, B.; Viladomat, F.; Codina, C.; Chem. Biodiversity 2011, 8, 115.
• ⁶
  Kreh, M.; Matusch, R.; Witte, L.; Phytochemistry 1995, 38, 773.
• ⁷
  Alberici, R. M.; Simas, R. C.; Sanvido, G. B.; Romão, W.; Lalli, P. M.; Benassi, M.; Cunha, I. B. S.; Eberlin, M. N.; Anal. Bioanal. Chem. 2010, 398, 265.
• ⁸
  Harris, G. A.; Galhena, S.; Fernández, F. M.; Anal. Chem. 2011, 83, 4508.
• ⁹
  Pereira, I.; Rodrigues, S. R. M.; Carvalho, T. C.; Carvalho, V. V.; Sanz Lobon, G.; Bassane, J. F.; Domingos, E.; Romão, W.; Augusti, R.; Vaz, B. G.; Anal. Methods 2016, 8, 6023.
• ¹⁰
  Klampfl, C. W.; Himmelsbach, M.; Anal. Chim. Acta 2015, 890, 44.
• ¹¹
  Snyder, D. T.; Schilling, M. C.; Hochwender, C. G.; Kaufman, A. D.; Anal. Methods 2015, 7, 870.
• ¹²
  Garrett, R.; Rezende, C. M.; Ifa, D. R.; Anal. Methods 2013, 5, 5944.
• ¹³
  Soparawalla, S.; Tadjimukhamedov, F. K.; Wiley, J. S.; Ouyang, Z.; Cooks, R. G.; Analyst 2011, 136, 4392.
• ¹⁴
  Mazzotti, F.; di Donna, L.; Taverna, D.; Nardi, M.; Aiello, D.; Napoli, A.; Sindona, G.; Int. J. Mass Spectrom. 2013, 352, 87.
• ¹⁵
  Deng, J.; Yang, Y.; Anal. Chim. Acta 2013, 785, 82.
• ¹⁶
  Muller, T.; Cooks, R. G.; Bull. Korean Chem. Soc. 2014, 35, 919.
• ¹⁷
  Schrage, M.; Shen, Y.; Claassen, F. W.; Zuilhof, H.; Nielen, M. W.; Chen, B.; van Beek, T. A.; J. Chromatogr. A 2013, 1317, 246.
• ¹⁸
  de Andrade, J. P.; Guo, Y.; Font-Bardia, M.; Calvet, T.; Dutilh, J.; Viladomat, F.; Codina, C.; Nair, J. J.; Zuanazzi, J. A. S.; Bastida, J.; Phytochemistry 2014, 103, 188.
• ¹⁹
  Pabuççuaglu, V.; Richomme, P.; Gözler, T.; Kivçak, B.; Freyer, A. J.; Shamma, M.; J. Nat. Prod. 1989, 52, 785.
• ²⁰
  Bastida, J.; Codina, C.; Viladomat, F.; Rubiralta, M.; Quirion, J. C.; Husson, H. P.; Ma, G. E.; J. Nat. Prod. 1990, 53, 1456.
• ²¹
  Suau, R.; Gomez, A. I.; Rico, R.; Phytochemistry 1990, 29, 1710.
• ²²
  Likhitwitayawuid, K.; Angerhofer, C. K.; Chai, H.; Pezzuto, J. M.; Cordell, G. A.; Ruangrungsi, N.; J. Nat. Prod. 1993, 56, 1331.
• ²³
  Bastida, J.; Lavilla, R.; Viladomat, F. In The Alkaloids; Cordell, G. A., ed.; Elsevier: Amsterdam, 2006, p. 87.
• ²⁴
  Portela-Cubillo, F.; Scott, J. S.; Walton, J. C.; J. Org. Chem. 2008, 73, 5558.
• ²⁵
  Lamoral-Theys, D.; Andolfi, A.; Van Goietsenoven, G.; Cimmino, A.; Le Calvé, B.; Wauthoz, N.; Mégalizzi, V.; Gras, T.; Bruyère, C.; Dubois, J.; Mathieu, V.; Kornienko, A.; Kiss, R.; Evidente, A.; J. Med. Chem. 2009, 52, 6244.
• ²⁶
  Osorio, E. J.; Berkov, S.; Brun, R.; Codina, C.; Viladomat, F.; Cabezas, F.; Bastida, J.; Phytochem. Lett. 2010, 3, 161.
• ²⁷
  Carvalho, T. C.; Oliveira, I. F.; Tose, L. V.; Vanini, G.; Kill, J. B.; Neto, A. C.; Machado, L. F.; Laboissiere, J. C.; Junior, V. L.; Vaz, B. G.; Romão, W.; Anal. Methods 2016, 8, 614.
• ²⁸
  Colletes, T. C.; Garcia, P. T.; Campanha, R. B.; Abdelnur, P. V.; Romão, W.; Coltro, W. K. T.; Vaz, B. G.; Analyst 2016, 141, 1707.
• ²⁹
  Oliveira, B. G.; Costa, H. B.; Ventura, J. A.; Kondratyuk, T. P.; Barroso, M. E. S.; Correia, R. M.; Pimentel, E. F.; Pinto, F. E.; Endringer, D. C.; Romão, W.; Food Chem. 2016, 204, 37.
• ³⁰
  Boit, H. G.; Döpke, W.; Naturwissenschaften 1960, 47, 109.
• ³¹
  de Andrade, J. P.; Pigni, N. B.; Torras-Claveria, L.; Berkov, S.; Codina, C.; Viladomat, F.; Bastida, J.; J. Pharm. Biomed. Anal. 2012, 70, 13.
• ³²
  de Andrade, J. P.; Pigni, N. B.; Torras-Claveria, L.; Guo, Y.; Berkov, S.; Reyes-Chilpa, R.; El Amrani, A.; Zuanazzi, J. A. S.; Codina, C.; Viladomat, F.; Bastida, J.; Rev. Latinoam. Quim. 2012, 40, 83.
• ³³
  Ibuka, T.; Irie, H.; Kato, A.; Uyeo, S.; Tetrahedron Lett. 1966, 39, 4745.
• ³⁴
  Kinstle, T. H.; Wildman, W. C.; Brown, C. L.; Tetrahedron Lett. 1966, 39, 4659.
• ³⁵
  Duffield, A. M.; Aplin, R. T.; Budzikiewicz, H.; Djerassi, C.; Murphy, C. F.; Wildman, W. C.; J. Am. Chem. Soc. 1965, 87, 4902.
• ³⁶
  Berkov, S.; Viladomat, F.; Codina, C.; Suárez, S.; Ravelo, A.; Bastida, J.; J. Mass Spectrom. 2012, 47, 1065.
• ³⁷
  Cédron, J. C.; Gutiérrez, D.; Flores, N.; Ravelo, A. G.; Estévez-Braun, A.; Bioorg. Med. Chem. 2012, 20, 5464.
• ³⁸
  De Angelis, G. G.; Wildman, W. C.; Tetrahedron 1969, 25, 5099.

Publication Dates

• Publication in this collection
  May 2017

History

• Received
  05 May 2016
• Accepted
  12 Aug 2016