Chemistry and Biological Activity of Alkaloids from the Genus Lycoris (Amaryllidaceae)

Lycoris Herbert, family Amaryllidaceae, is a small genus of about 20 species that are native to the warm temperate woodlands of eastern Asia, as in China, Korea, Japan, Taiwan, and the Himalayas. For many years, species of Lycoris have been subjected to extensive phytochemical and pharmacological investigations, resulting in either the isolation or identification of more than 110 Amaryllidaceae alkaloids belonging to different structural types. Amaryllidaceae alkaloids are frequently studied for their interesting biological properties, including antiviral, antibacterial, antitumor, antifungal, antimalarial, analgesic, cytotoxic, and cholinesterase inhibition activities. The present review aims to summarize comprehensively the research that has been reported on the phytochemistry and pharmacology of the genus Lycoris.


Introduction
Plants of the Amaryllidaceae family, which consists of about 85 genera and 1100 species, are distributed over the tropical and warm regions of the world. They have been extensively used as folk medicines to treat various diseases in many countries and areas [1][2][3]. Chemically, the Amaryllidaceae family is known for its unique alkaloid constituents, named Amaryllidaceae alkaloids (AAs), which display a wide range of biological activities including acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE) inhibition effects, and antitumor, antifungal, antibacterial, antiviral, and antimalarial properties [1,[4][5][6]. The most known representative of AAS is galanthamine, which is currently used for the treatment of early and intermediate states of Alzheimer's disease (AD) [7].
The present review summarizes phytochemical studies carried out on the genus Lycoris, focusing on the occurrence, isolation, identification and biological activities of its alkaloids. Lycoris species have been used for a long time in traditional medicine.

Genus Lycoris: Occurrence, Ethnobotany
The genus Lycoris Herbert consists of about 20 species which are distributed in the moist warm temperate woodlands of eastern Asia, as in China, Korea, Japan, Taiwan, and the Himalaya [8,9]. The genus was established by Herbert in 1821 [10], and L. aurea has been assigned as the type species. In the mid-19th to early 20th centuries, nine new species were published by various European authors (L. africana, L. straminea, L. sewerzowii, L. squamigera, L. sanguinea, L. terraccianii, L. sprengeri, L. incarnata and L. argentea). In the first half of the 20th century three new species from Japan were described (L. albiflora, L. koreana and L. kiusiana) [10]. After that, American botanist Hamilton P. Traub recorded 10 new taxa (two with Moldenke) of Lycoris, based mainly on materials introduced from China, Japan and cultivated

Phytochemistry of the Genus Lycoris
Of the accepted 23 Lycoris species, fourteen have been chemically investigated (Table 1). Major attention within the reported phytochemical studies has been given to the study of alkaloids, since they are the most studied constituents of this genus, and only little attention has been given to other components. The AAs are largely restricted to the family Amaryllidaceae, specifically the subfamily Amaryllidoideae [20]. They are derived from the aromatic acids phenyalanine and tyrosine, which are used to produce key intermediates in the biosynthesis of the AA 4 -O-methylnorbelladine [21,22]. According to the name of this key intermediate, this biosynthetic pathway of AAs is called the norbelladine pathway [23]. Recently, several review articles provide detailed coverage of the biosynthesis of AAs [1,[23][24][25][26], and thus we will not repeat this in the current review. Altogether, 116 AAs of various structural types have been isolated in either pure form or identified by different analytical methods (e.g., GC-MS or HPLC-MS) in the studied Lycoris plants (Table 1; Figures 1 -5). The reported alkaloids belong to the belladine (1), crinine (2-6), galanthamine (7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21), galanthindole (22), haemanthamine (23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34), homolycorine , hostasinine (60), ismine (61), lycorine (62-92), montanine (93-98), narciclasine (99-107) and tazettine (108-116) structural types (      The most studied Lycoris species, L. radiata, also called spider lily, is a horticultural plant widely distributed in the south of China, Vietnam, Korea, Nepal and Malaysia [14,15,27]. Altogether 79 Aas have been either identified or isolated from either the bulbs or flowers of this plant (Table 1). This species is rich in galanthamine-, haemanthidine-, homolycorine-and lycorine-type Aas. From the bulbs of L. radiata, the alkaloid colchicine (128) has recently been isolated; this is also derived from tyrosine, as are AAs, but it is metabolite typical of the genus Colchicum [28]. The question is whether it is a real product of the Lycoris plant, and not a product of some contamination with another plant, since this is the only report of this alkaloid-type within the whole Amaryllidaceae family. The same is also true for the reports of further alkaloids of other structural types (117-120 and 122-128; Table 1). The natural existence of an N-chloromethyl moiety in galanthamine-, lycorine-, and montanine-type AAs, such as in N-chloromethylgalanthamine (12), N-chloromethyllycoramine (16), N-chloromethylungiminorine (66), N-chloromethylnarcissidin (67), and lycolongirine C (98) isolated from bulbs of L. longituba, L. radiata, and L. sprengeri must be reinvestigated, because when halogenated solvents are used during the isolation process, this can result in the formation of N-chloromethyl alkaloid derivative artifacts [29]. In the course of phytochemical study of the mentioned Lycoris species, dichloromethane has been used for extraction and column chromatography of the alkaloids, and this can be the explanation for the isolation of the mentioned compounds [30]. Either HPLC/MS or GC/MS analysis should be used for analysis of the alkaloidal extract before the separation process. Then, the presence of isotopes of 35 Cl and 37 Cl in the mass spectra of the separated compounds can indicate the natural presence of N-chlormethylated derivatives. To avoid the possible formation of the mentioned artifacts, ethylacetate is recommended to be used for the preparation of the alkaloidal extract instead of halogenated solvents such as dichloromethan or chloroform. Another question is the natural presence of the butoxy-moiety at the C6 position in homolycorine-type compounds such as 2α-hydroxy-6-O-n-butyloduline (38) and O-n-butyllycorenine (50), both isolated from L. aurea in a study from 2014, due to the reactivity of the carbonyl and hemiacetal groups [29]. The authors stated the use of 95% alcohol for the total extraction of fresh bulbs, but the exact type of alcohol was not specified at the beginning of the isolation process [31]. It can be assumed that butanol was used for this initial isolation step. To avoid discussion and speculation as to whether the isolated product is really a natural compound and not an isolation artefact, it is necessary to specify the solvents used within all steps of the phytochemical study.
Crude extract, DCME extract and pure alkaloids isolated from bulbs of L. aurea were also evaluated for their antiproliferative activities against the SH-SY5Y cell line. The crude and DCME extracts revealed cytotoxicity at concentrations of 5 µg/mL [31]; of the pure alkaloids, only lycorine (80) demonstrated significant cytotoxicity at a concentration 6.25 µM [31].
Alkaloidal extracts of three Lycoris species were screened for their cytotoxic potential against HepG2 cells at a concentration of 10 µg/mL with inhibitory rates of 78.0%, 84.9%, and 66.8% for L. aurea, L. radiata and L. guangxiensis, respectively [32].
An alcoholic extract of fresh bulbs of L. albiflora showed promising cytotoxic activity against HL-60 cells, with an IC 50 value of 1.7 µg/mL [39]. This resulted in a detailed phytochemical study being undertaken to isolate pure AAs and test them for their cytotoxic activity; fifteen AAs were isolated (Table 1), which were tested for their cytotoxic activity against the cancerous cell line HL-60. The most potent AAs were also subjected to a cytotoxic screening against HSC-2 cells [39]. Narciclassine-type alkaloids 7-deoxynarciclasine (99; also known as lycoricidine) and narciclasine (100, also known as lycoricidinol) induced apoptosis in both HL-60 and HSC-2 cells. Moreover, narciclasine (100) induced transient autophagy and morphological changes in mitochondria in the early stages of the apoptotic cell death process in HSC-2 cells [39]. Within previous studies, narciclasine (100) exhibited potent in vitro cytotoxicity against various cancer cells and showed great potential against primary brain cancers, as well as brain metastases in vivo [63][64][65]. Within the latest study, narciclasine (100) displayed preferential cytotoxicity towards primary effusion lymphoma cell lines (PEL), an aggressive type of non-Hodgkin lymphoma, with IC 50 values ranging from 7 to 14 nM [66]. 7-Deoxynarciclasine (99) displayed approximately 10 times lower cytotoxicity against the tested PEL cell lines (IC 50 = 82-162 nM) [66]. Previous in vitro studies frequently focused on the cytotoxicity of narciclasine against fibroblast (IC 50 = 7.5 µM) and cancer cells (IC 50 = 30 nM), which indicated the compound's selectivity to cancer cells and only higher concentrations affected the viability of fibroblasts [67]. On the other hand, it was reported that narciclasine (100) showed only modest anti-tumor effects in mice in vivo, with considerable toxicity [68]. Thus, narciclasine (100) has not been tested in human clinical trials up to now. The inhibitory effects on L02 (human normal liver cell line) and murine macrophages RAW264.7 indicated that narciclasine (100) might have significant side effects, and, therefore, further studies are urgently needed [16]. Narciclasine (100) has also been shown to inhibit the cytotoxicity of calprotectin in rat adjuvant arthritis mode, and several studies have reported that narciclasine (100) exhibits strong anti-inflammation activity in vitro and in vivo [69]. LPS-stimulated RAW264.7 cells were employed to investigate the anti-inflammatory effects of narciclasine (100) in order to explore its underlying mechanism [16]. Recently, narciclasine was named 'Molecule of the Week' by the American Chemical Society (ACS) for its potential as a cancer drug [66].

Antimalarial Activity
Malaria is one of the most common vector-borne infectious diseases. This disease is caused by protozoan parasites of the genus Plasmodium [77]. Alkaloids isolated from L. radiata were evaluated in vitro for their antimalarial activity using the drug-resistant D-6 strain and drug sensitive W-2 strain of P. falciparum [47]. Within the tested AAs, only 5,6-dehydrolycorine (75) exhibited antimalarial activity, with IC 50 values of 2.3 µM for the D-6 strain and 1.9 µM for the W-2 strain of P. falciparum [47]. Other studied AAs displayed only weak or no antimalarial activity against the studied strains (Table 4).
Aphids are one of the most destructive and economically important pests of plants on earth and extensive use of insecticides has resulted in the development of insecticide resistance among aphids across regions [78]. Thus, the insecticidal activity of ten AAs isolated from L. radiata against Aphis citricola has been studied [28]. LD 50 values were measured by a capillary drip method and nine of the tested AASs displayed aphicial activity. N-Allylnorgalanthamine (10) possessed the highest aphicial activity (LD 50 = 4.92 ± 0.83 ng/aphid), which was comparable with the commercial pesticide methomyl (LD 50 = 2.91 ± 0.18 ng/aphid). Inhibition of AChE is a main target enzyme for many insecticides (especially carbamates and organophosphates) [79]; the in vitro inhibition of AChE of N-allylnorgalanthamine (10) has also been studied. This compound demonstrated remarkable inhibition activity against AChE in A. citricola with a value of IC 50 = 2.1 nM.

Conclusions
In conclusion, this review summarizes the ethnobotanical, phytochemical, and pharmacological information about plants and AAs of the genus Lycoris. So far, fourteen Lycoris species have been phytochemically studied, and nearly 120 AAs of different structural types have been either identified or isolated. The presence and structures of some reported AAs must be reevaluated, as they may be isolation artifacts, and not naturally occurring compounds. Lycoris plants are above all a rich source of homolycorine-and lycorine-type AAs. Most of the isolated AAs have been studied for different biological activities with impact on antitumor and neuroprotective activities. The antimalarial, antiviral and insecticidal activities of some AAs have also been described. In the light of the presented overview of scientific data, the genus Lycoris can be recognized as an interesting source of different structural types of AAs with a wide range of biological activities.