Isolation and Biological Characterization of Homoisoflavanoids and the Alkylamide N-p-Coumaroyltyramine from Crinum biflorum Rottb., an Amaryllidaceae Species Collected in Senegal

Crinum biflorum Rottb. (syn. Crinum distichum) is an Amaryllidaceae plant used in African traditional medicine but very few studies have been performed on this species from a chemical and applicative point of view. Bulbs of C. biflorum, collected in Senegal, were extracted with ethanol by Soxhlet and the corresponding organic extract was purified using chromatographic methods. The pure compounds were chemically characterized by spectroscopic techniques (1D and 2D 1H and 13C NMR, HR MS and ECD) and X-ray analysis. Four homoisoflavonoids (1–4) and one alkylamide (5) were isolated and characterized as 5,6,7-trimethoxy-3-(4-hydroxybenzyl)chroman-4-one (1), as 3-hydroxy-5,6,7-trimethoxy-3-(4-hydroxybenzyl)chroman-4-one (2), as 3-hydroxy-5,6,7-trimethoxy-3-(4-methoxybenzyl)chroman-4-one (3) and as 5,6,7-trimethoxy-3-(4-methoxybenzyl)chroman-4-one (4), and the alkylamide as (E)-N-(4-hydroxyphenethyl)-3-(4-hydroxyphenyl)acrylamide (5), commonly named N-p-coumaroyltyramine. The relative configuration of compound 1 was verified thanks to the X-ray analysis which also allowed us to confirm its racemic nature. The absolute configurations of compounds 2 and 3 were assigned by comparing their ECD spectra with those previously reported for urgineanins A and B. Flavanoids 1, 3 and 4 showed promising anticancer properties being cytotoxic at low micromolar concentrations towards HeLa and A431 human cancer cell lines. The N-p-coumaroyltyramine (5) was selectively toxic to A431 and HeLa cancer cells while it protected immortalized HaCaT cells against oxidative stress induced by hydrogen peroxide. Compounds 1–4 also inhibited acetylcholinesterase activity with compound 3 being the most potent. The anti-amylase and the strong anti-glucosidase activity of compound 5 were confirmed. Our results show that C. biflorum produces compounds of therapeutic interest with anti-diabetic, anti-tumoral and anti-acetylcholinesterase properties.


Introduction
Plants and microorganisms are well-known sources of bioactive metabolites which have only been partly investigated [1]. Among the plants' kingdom, the Amaryllidaceae is

Plant Material
Bulbs of C. biflorum were collected in Senegal, in Kaffrine department, in December 2018. A senior scientist from the Herbarium of IFAN of University Cheikh Anta Diop of Dakar taxonomically identified the plant materials.

Crystal Structure Determination of Compound 1
Single crystals of 1 suitable for X-ray structure analysis were obtained by slow evaporation of a CHCl 3 -isoPrOH 9:1 v/v solution. One selected crystal was mounted at ambient temperature on a Bruker-Nonius KappaCCD diffractometer (Bruker-Nonius, Delft, The Netherlands) (graphite monochromated MoKα radiation, λ = 0.710 73 Å, CCD rotation images, thick slices and ϕ and ω scan to fill the asymmetric unit). A semi-empirical absorption correction (multiscan, SADABS) was applied. The structure was solved by direct methods using the SIR97 program [34] and anisotropically refined by the full-matrix least-squares method on F 2 against all independently measured reflections using the SHELXL-2018/3 program [35] with the aid of program WinGX [36]. Water solvent crystallization molecules are present in the structure. The hydroxy and water H atoms were located in different Fourier maps and freely refined with Uiso(H) equal to 1.2 Ueq of the carrier atom. All of the other hydrogen atoms were introduced in calculated positions and refined according to the riding model with C−H distances in the range of 0.93−0.96 Å and with Uiso(H) equal to 1.2 Ueq or 1.5 Ueq (Cmethyl) of the carrier atom. One stereogenic center is present in the compound that crystallizes in the centrosymmetric P-1 space group as a racemate. The E-statistics indicate that the structure is centrosymmetric. Two independent X-ray structure analyses performed on different crystals confirmed the result. Unitary cell parameters were checked on several crystals. Figures were generated using ORTEP-3 [36] and Mercury-CSD-3.9. [37]. Crystallographic data of 1: empirical formula: C 19 H 20 O 6 ·H 2 O; formula weight: 362.36 g mol −1 ; triclinic, P-1; a: 8.368(2) Å; b: 10.183(2) Å; c: 12.0420(6) Å; α: 104.470 (8) • ; β: 108.252 (13) • ; γ: 100.010 (19) • ; V: 907.4(3) Å 3 ; Z: 2, Dx: 1.326 Mg/m3. All homoisoflavanoid crystallographic data for (1) were deposited in the Cambridge Crystallographic Data Centre with deposition number CCDC 2092030. These data can be obtained free of charge from www.ccdc.cam.ac.uk/data_request/cif.

MTT Assay
Cell viability was assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay following the published procedure [40]. Briefly, cells were seeded at Biomolecules 2021, 11, 1298 5 of 21 10 5 /cm 2 density in 96-well plates. Twenty-four hours later, the medium was changed and supplemented with the specified concentrations of metabolite (from 0.5 to 10 mM in DMSO) for 24 and 48 h. MTT solution 1:10 (stock solution 5 mg/mL) was added to each well and the absorbance was measured in dual-wavelength mode (570 nm and 630 nm). The percentage of cell viability was calculated as follows: mean (A570-A630) and compared to cells supplemented with DMSO alone. Values shown in the plot are mean ± SD of sixfold determinations. Mean and the standard deviation was calculated on biological triplicates using GraphPad Prism8 software (GraphPad, San Diego, CA, USA).

Detection of DNA Damage
Cells were seeded in 35 mm dishes on micro cover glasses (BDH) and treated with the metabolite at a concentration of 10 µM. At 48 h after treatment, cells were washed with cold phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde Sigma-Aldrich (Merck Life Science, Milan, Italy) for 15 min at RT. Cells were permeabilized with ice-cold 0.5% Triton X-100 for 5 min and then washed with PBS. Cells were then incubated with phospho-histone H2A.X (Ser139) antibody (from Cell Signaling Technologies 9542, Boston, MA, USA) for 1 h, followed by DAPI (Sigma-Aldrich) for 3 min and washed with PBS/0.05% Tween. Coverslip was mounted with Ibidi mounting medium (Ibidi GmbH, Martinsried, Germany). Images were taken with a Zeiss confocal laser-scanning microscope Axio Observer (Zeiss, Ostfilden, Germany) (scale bar, 20 µm). A 40× objective was used and image analysis was performed using Fiji ImageJ open source software project (https://imagej.net/imaging/). All the images were taken with the same setting [39].

Western Blot Analysis
Western blot was performed as previously reported [39,41]. Briefly, 30 µg of whole-cell extracts were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), subjected to Western blot and incubated overnight at 4 • C with antibodies. Antibodies against p21WAF, Poly [ADP-ribose] polymerase 1 (PARP1) and actin were from Cell Signaling Technologies 9542, Boston, MA, USA. Each experiment was run in triplicate. Signal intensities of Western blot bands were quantified by Quantity One analysis software (Version Number 2, Biorad Laboratories, London, UK) and analyzed by GraphPad Prism 8.0.2 software (GraphPad, San Diego, CA, USA).

DCFDA Assay
N-p-coumaroyltyramine antioxidant activity was measured using 2 −7 dichlorofluorescein diacetate (DCFDA), a non-fluorescent compound permeable to the cell membrane, which can be oxidized by reactive oxygen species (ROS) giving a fluorescent compound. Cells were seeded at 2.5 × 10 4 in 96 well and pre-treated with N-p-coumaroyltyramine (10 and 100 mM). The medium was removed after 4 h and 1 mM (3%) H 2 O 2 was added for 45 min, 1.5 and 2.0 h. Cells were washed with PBS and a fresh medium with DCFDA (30 mM) was added for 45 min, then DCFDA was removed by washing in PBS and the cells were harvested. The measurement of ROS was obtained using the Sinergy H4 microplate reader (Gen5 2.07, Thermofisher, Waltham MA, USA). The fluorescence emitted from the cells treated with DCFDA was compared to the untreated cells. Trolox was used as a positive control. Values shown in the plot are mean ± SD of sixfold determinations. The mean and the standard deviation were calculated on biological triplicates using GraphPad Prism 8.0.2 software (GraphPad, San Diego, CA, USA).

2Pseudotyped HIV-1 GFP Infectivity Assay
The anti-HIV-1 activity of compounds 1-5 was evaluated using VSV-G pseudotyped NL43 GFP infection of human monocytic THP-1 cells. THP-1 and NL4-3 GFP were generously provided by Lionel Berthoux and Amita Singh and are described in Ka et al. (2021) [13]. THP-1 cells were seeded at 2.0 × 10 4 cells per well in 96 well-plates. The next day, cells were treated with 4 concentrations of each compound (12.5, 25, 50 and 100 mM) and then infected with HIV GFP at a MOI of 1. After 72 h, cells were stained with propidium iodide (PI, 0.5 mg/mL) and both PI + and HIV-1 GFP + infected cells frequencies were assessed on a FC500 MPL cytometer (Beckman Coulter, Inc., Mississauga, ON, Canada) and analyzed using FlowJo software (FlowJo LLC, BD Biosciences, Ashland, OR, USA). Matched concentrations of dimethyl sulfoxide (DMSO) were used as negative controls. All infection assays were performed in triplicate.
2.12. α-Glucosidase and α-Amylase Inhibitor Assay α-glucosidase and α-amylase inhibitor screening kits (colorimetric) were purchased from Biovision (Milpitas, CA, USA). In total, 10 mM of stock solution of all the tested compounds were dissolved in DMSO and serially diluted in the assay buffer of each kit. Experiments were performed according to the manufacturer's protocol. Briefly, for the α-glucosidase assay, 10 µL of serially diluted compounds at the corresponding concentration (10 nm-1 mM) were added into designated wells of clear 96 well-plates. Subsequently, 10 µL of the α-glycosidase enzyme was added to each well and volume was adjusted to 80 µL and plates were incubated for 15-min at room temperature in dark condition. Then, 20 µL of α-glycosidase substrate mixture was added in all wells and kinetic of reaction was measured at OD: 410 nm for 60 min at 2 min intervals by using a multiplate reader, Biotek instrument, Inc., Canada. Enzyme control (no inhibitor), background control (no enzyme), solvent control (DMSO) and inhibitor control (acarbose) were included in the plates. For the α-amylase assay, 50 µL of serially diluted compounds (3.25 µM to 500 µM) were added into a clear 96-well plate with 50 µL of assay buffer and 50 µL of α-amylase enzymes. The plate was incubated at room temperature in the dark for 10 min. Then 50 µL of the α-amylase substrate was added in all wells. The kinetic of reaction was measured at OD:410 nm for 26 min at intervals of 2 min by using a multiplate reader. Control αamylase inhibitor was provided by the manufacturer, enzyme control, background control and solvent control were all included. Enzyme inhibition was calculated according to Zhang et al. (2014) [42]. In summary, ODs were plotted according to the time for each sample. Areas under the curve (AUC) were calculated, and enzyme inhibition was measured as 100−(AUC compound /AUC enzyme ) × 100 for each dilution of each compound.

Anti-Acetylcholinesterase Assays
In vitro acetylcholinesterase (AChE) activity was assessed exactly as in Ka et al. (2020) [12] following Ellman's colorimetric protocol [43] with the Acetylcholinesterase Assay Kit (Abcam Inc., Boston, MA, USA). Briefly, 50 µL serial dilutions (3.9-500 µM) of compounds 1-5 were prepared in Tris-HCl pH = 7.9 buffer into designated wells of a clear 96 well-plate. A total of 5 µL of DTNB was added in each well, then 50 µL of diluted acetycholinesterase was added. The plate was incubated for 10 min in the dark. Matched concentrations of DMSO were used as a negative control. Kinetic of reaction was measured in a multiple plate reader at 410 nm in kinetic mode for 40 min at room temperature. The percentage of anti-AChE inhibition was calculated according to the following formula: 100 − [((E − S)/E) × 100], where E is the activity of the enzyme with matched concentrations of DMSO and S is the activity of the enzyme with the test sample.

Statistical Analysis
Statistical analyses were carried out using the GraphPad Prism version 8.1.2 (https://www.graphpad.com/scientific-software/prism/). Data were represented as the mean ± standard deviation and analyzed for statistical significance using ordinary oneway analysis of variance (ANOVA) and multiple comparisons. For all tests, p < 0.05 was considered to indicate a statistically significant difference.

Identification of Metabolites Isolated from C. biflorum
The purification of the crude organic extract from bulbs of C. biflorum allowed us to isolate four homoisoflavanoids ((1)-(4), Figure 1) identified using spectroscopic (essentially 1D and 2D 1 H and 13 C NMR and HR MS) methods as 5,6,7-trimethoxy-3-(4hydroxybenzyl)chroman-4-one (1), as 3-hydroxy-5,6,7-trimethoxy-3-(4-hydroxybenzyl)chroman-4-one (2), as 3-hydroxy-5,6,7-trimethoxy-3-(4-methoxybenzyl)chroman-4-one (3) and as 5,6,7-trimethoxy-3-(4-methoxybenzyl)chroman-4-one (4). The identification of 1 was performed comparing its 1 H NMR spectrum with that reported in the literature when it was isolated for the first time from Scilla nervosa, subsp. rigidifolia collected in Botswana where it is used in Zulu folk medicine to treat rheumatic fever and as a purge for children [30]. crystallizes in the P-1 space group with one molecule of 1 and one H 2 O solvent molecule contained in the independent unit. All bond lengths and angles are in a normal range. The molecule consists of a substituted cromanone system. The six-membered croman-4-one ring assumes an envelope conformation with C2 atom at the flap. The 4-hydroxybenzyl substituent at C3 is in the equatorial position and places nearly parallel to the benzene ring of the chromanone system (Figure 2A,B). The crystal packing is stabilized by strong OH . . . O hydrogen bonds involving 1 and solvent water molecules ( Figure 3A,B). The homoisoflavonoid (1) molecule contains one stereogenic center at C3 atom and crystallizes in the centrosymmetric P-1 space group as a racemate. The racemic nature of crystals was confirmed by performing two independent X-ray structure analyses on different crystals ( Figure 2B). This result can be ascribed to easy inversion at the chirality center, because of the presence at C-3 of a proton which could exchange by keto-enol tautomerism. The ECD spectrum of 1 ( Figure 4, black line) cannot be interpreted due to the presence of a racemic mixture. These results were confirmed by the optical inactivity found by measuring the specific optical rotation of compound 1. The homoisoflavonoid (1) molecule contains one stereogenic center at C3 atom and crystallizes in the centrosymmetric P-1 space group as a racemate. The racemic nature of crystals was confirmed by performing two independent X-ray structure analyses on different crystals ( Figure 2B). This result can be ascribed to easy inversion at the chirality center, because of the presence at C-3 of a proton which could exchange by keto-enol tautomerism. The ECD spectrum of 1 ( Figure 4, black line) cannot be interpreted due to the presence of a racemic mixture. These results were confirmed by the optical inactivity found by measuring the specific optical rotation of compound 1. The homoisoflavonoid (1) molecule contains one stereogenic center at C3 atom and crystallizes in the centrosymmetric P-1 space group as a racemate. The racemic nature of crystals was confirmed by performing two independent X-ray structure analyses on different crystals ( Figure 2B). This result can be ascribed to easy inversion at the chirality center, because of the presence at C-3 of a proton which could exchange by keto-enol tautomerism. The ECD spectrum of 1 ( Figure 4, black line) cannot be interpreted due to the presence of a racemic mixture. These results were confirmed by the optical inactivity found by measuring the specific optical rotation of compound 1.
Homoisoflavanoides 2 and 3 were identified by comparing their physic (ECD) and spectroscopic data ( 1 H NMR) with those reported in the literature for urgineanins B and A, isolated from Urginea depressa. This is an Asparagaceae collected in South Africa and used for its antiproliferative activity against the A2780 ovarian cancer cell line [31].  [31] which assigned a R configuration at C-3 of the two homoisoflavanoids by comparison of their experimental ECD spectra with those reported in literature for caesalpiniaphenol A [31].
1 with hydrogen bond pattern drawn as cyan and red lines. Ball-and-stick style.
The homoisoflavonoid (1) molecule contains one stereogenic center at C3 atom and crystallizes in the centrosymmetric P-1 space group as a racemate. The racemic nature of crystals was confirmed by performing two independent X-ray structure analyses on different crystals ( Figure 2B). This result can be ascribed to easy inversion at the chirality center, because of the presence at C-3 of a proton which could exchange by keto-enol tautomerism. The ECD spectrum of 1 (Figure 4, black line) cannot be interpreted due to the presence of a racemic mixture. These results were confirmed by the optical inactivity found by measuring the specific optical rotation of compound 1.  Homoisoflavanoides 2 and 3 were identified by comparing their physic (ECD) and spectroscopic data ( 1 H NMR) with those reported in the literature for urgineanins B and A, isolated from Urginea depressa. This is an Asparagaceae collected in South Africa and used for its antiproliferative activity against the A2780 ovarian cancer cell line [31].  The fourth homoisoflavanoid (4) was identified as the 5,6,7-trimethoxy-3-(4-methoxybenzyl)chroman-4-one by comparing its physic and spectroscopic data with those reported for the trimethyl derivative of 3,9-dihydroautumnalin obtained together to other homoisoflavanoids from Eucomis autumnalis (Liliaceae) [32]. The same derivative was also obtained from 3,9-dihydroeucomnsalin, which resulted to be identical to 3,9-dihydroautumnalin, which was isolated from the bulbs of Muscari comsum (commonly named lampascioni) collected in Basilicata region, Italy, where they are used in traditional cuisine as bitter plants [44]. The identification of compound 4 was also supported by ESI MS data which showed the sodiated [M + Na] + and protonated [M + H] + adduct ions at m/z 381 and 359. The ECD spectrum of 4 ( Figure 4, red line) cannot be interpreted, as in the case of compound 1, due to the presence of a racemic mixture, as confirmed by the optical inactivity found by measuring the specific optical rotation of 4.
From the ethanolic extract of the C. biflorum bulbs and alkylamide was also isolated The fourth homoisoflavanoid (4) was identified as the 5,6,7-trimethoxy-3-(4-methoxybenzyl) chroman-4-one by comparing its physic and spectroscopic data with those reported for the trimethyl derivative of 3,9-dihydroautumnalin obtained together to other homoisoflavanoids from Eucomis autumnalis (Liliaceae) [32]. The same derivative was also obtained from 3,9-dihydroeucomnsalin, which resulted to be identical to 3,9-dihydroautumnalin, which was isolated from the bulbs of Muscari comsum (commonly named lampascioni) collected in Basilicata region, Italy, where they are used in traditional cuisine as bitter plants [44]. The identification of compound 4 was also supported by ESI MS data which showed the sodiated [M + Na] + and protonated [M + H] + adduct ions at m/z 381 and 359.
The ECD spectrum of 4 ( Figure 4, red line) cannot be interpreted, as in the case of compound 1, due to the presence of a racemic mixture, as confirmed by the optical inactivity found by measuring the specific optical rotation of 4.
From the ethanolic extract of the C. biflorum bulbs and alkylamide was also isolated and identified compound 5, or (E)-N-(4-hydroxyphenethyl)-3-(4-hydroxyphenyl)acrylamide (5, Figure 1). Compound 5 was identified by comparison of its spectroscopic data with those reported in the literature when isolated from the first time from crude Chinese drug "Xiebai", the tuber of Allium bakeri Reg. (Liliaceae) and used for inhibition on human platelet aggregation [45]. A similar comparison was done with the data reported when 5 was isolated together with some alkaloids from Fumaria indica, collected in Multan City (Punjab, Pakistan) and inappropriately indicated as an alkaloid instead of amide [33]. The identification of compound 5 was also supported by ESI MS data which showed the sodiated [M + Na] + adduct ion at m/z 306. When the same spectrum was recorded in negative modality, the pseudomolecular ion [M − H] − at m/z 282 was observed.

In Vitro Cytotoxicity
Considering that the anti-tumor activities of homoisoflavonoids (1-4) from C. biflorum have not yet been addressed, we evaluated the in vitro cytotoxicity of isolated homoisoflavonoids towards HaCaT, A431 and HeLa human cell lines using the MTT viability assay. The Np-coumaroyltyramine (5) alkylamide was also included in these experiments. Cells were plated and supplemented with each metabolite, at low micromolar concentrations ranging from 0.5 to 10 µM in DMSO, and incubated at 37 • C for 24 and 48 h. As shown in Figure 6, all tested metabolites were toxic for cancer cell lines, in a dose and time-dependent way, with HeLa cells being more sensitive than A431. Remarkably, 48 h treatment with 5 µM metabolite 1 reduced HeLa cells viability to less than 20% of the untreated control. All metabolites were less active on HaCaT immortalized keratinocytes at 24 h of incubation over the range of concentration tested although a significant reduction of HaCaT cell viability was caused by metabolites 3 and 4 when the incubation was extended for 48 h. Table 1 lists the IC 50 values obtained with the test compounds are means of triplicates at 24 h. The N-p-coumaroyltyramine (5) was also found to be toxic on A431 and HeLa cancer cells while at the lower concentration tested, 0.5 and 1 µM, it was shown to significantly increase HaCaT cell viability ( Figure 6). Detection of nuclear γ-H2A.X foci provides indirect evidence of the occurrence of DNA double-strand breaks (DSB) and/or DNA replication stress [40]. Upon induction of a DNA double-strand break, the H2A.X histone becomes rapidly phosphorylated at serine 139 to form γH2AX [46]. This phosphorylation event is dynamic, complex and depends on interactions between MDC1, H2AX and ATM and other kinases to persist [47]. This amplified response is easily detected using specific antibodies against γ-H2AX, manifesting discrete nuclear foci.   The formation of γ-H2AX foci by immunofluorescence using the antibody against the histone H2A.X phosphorylated in Serine 139 was monitored considering that the homoisoflavonoids isolated from C. biflorum reduced cell viability. HaCaT, HeLa and A431 cells were treated with 10 µM metabolites 3 and 4 for 24 h to detect DNA damage foci. As shown in Figure 7 a remarkable increase of nuclear γH2AX foci was observed in all tested cell lines.
1 IC50 was calculated after 24 h of incubation.
The formation of γ-H2AX foci by immunofluorescence using the antibody against the histone H2A.X phosphorylated in Serine 139 was monitored considering that the homoisoflavonoids isolated from C. biflorum reduced cell viability. HaCaT, HeLa and A431 cells were treated with 10 μM metabolites 3 and 4 for 24 h to detect DNA damage foci. As shown in Figure 7 a remarkable increase of nuclear γH2AX foci was observed in all tested cell lines. Determination of reactive oxygen species (ROS) induced by H2O2 in HaCaT cells treated with 10 and 100 µ M N-p-coumaroyltyramine alkylamide revealed that 5 has antioxidant activity comparable to Trolox, the water-soluble derivative of vitamin E used as a positive control (Figure 8). Remarkably, treatment of HaCaT cells with 10 µ M metabolite 3 or 4 did not increase the production of reactive oxygen species (ROS) implying that ROS generation was not responsible for the observed DNA damage ( Figure 9A,B).

Determination of reactive oxygen species (ROS) induced by H 2 O 2 in
HaCaT cells treated with 10 and 100 µM N-p-coumaroyltyramine alkylamide revealed that 5 has antioxidant activity comparable to Trolox, the water-soluble derivative of vitamin E used as a positive control (Figure 8). Remarkably, treatment of HaCaT cells with 10 µM metabolite 3 or 4 did not increase the production of reactive oxygen species (ROS) implying that ROS generation was not responsible for the observed DNA damage ( Figure 9A,B). Determination of reactive oxygen species (ROS) induced by H2O2 in HaCaT cells treated with 10 and 100 µ M N-p-coumaroyltyramine alkylamide revealed that 5 has antioxidant activity comparable to Trolox, the water-soluble derivative of vitamin E used as a positive control (Figure 8). Remarkably, treatment of HaCaT cells with 10 µ M metabolite 3 or 4 did not increase the production of reactive oxygen species (ROS) implying that ROS generation was not responsible for the observed DNA damage ( Figure 9A,B).

Antiviral Activity
Flavonoids and alkylamides have also been shown to exhibit antiviral activity [49]. Thus, we tested the anti-retroviral effect of compounds 1-5 using VSV-G pseudotyped HIV-1 particles. Interestingly, all compounds displayed anti-HIV-1 potential at 100 µM ( Figure 12A). Infection dropped from a mean of 8.16%, when cells were treated with DMSO, to means of 3.78, 4.03, 2.51, 2.76 and 2.12 %, when compounds 1 to 5 were added, respectively. These differences were statistically significant for compounds 3, 4 and 5. At 50 µM, there was a 2.0, 2.4 and 1.6-fold decrease in infection when 3, 4 and 5 were added to the cell media, respectively. At lower concentrations only 3 and 4 decreased infection levels. Cytotoxicity was verified following propidium iodide staining of THP-1 cells after an incubation of 72 h with compounds ( Figure 12B). Compounds 1, 2 and 5 were weakly cytotoxic to THP-1 at 100 µM with 8.74, 4.15 and 5.97% of PI + cells, respectively. In total, 30-40% of cells treated with 3 and 4 were dead at all concentrations tested. Thus, although cytotoxicity could contribute to the observed antiviral activity, this experiment was performed using a single-cycle infection system, lessening the effect of cell loss on further cycles of viral replication. In conclusion, C. biflorum contains compounds with interesting anti-retroviral activity.
. Plants and microorganisms are well-known sources of bioactive metabolites which have only been partly investigated [1]. Among the plants' kingdom, the Amaryllidaceae is a plant family extensively studied essentially for its alkaloids and related isocarbostiryls content which show a broad spectrum of biological activities [2][3][4][5]. These plants are principally diffused in tropical and subtropical regions of the world, as Andean South America, the Mediterranean basin and Southern Africa [6], and include ca 1600 species classified into about 75 genera [7]. Hundreds of Amaryllidaceae alkaloids with different structures and biological activities were isolated and reported in several reviews. A Special Issue of the journal Molecules was edited by Bastida J. and Berkov S. in 2020 on different  [12]) was used as a positive control. For all assays, enzyme, no compounds and inhibition controls were included.

Antiviral Activity
Flavonoids and alkylamides have also been shown to exhibit antiviral activity [49]. Thus, we tested the anti-retroviral effect of compounds 1-5 using VSV-G pseudotyped HIV-1 particles. Interestingly, all compounds displayed anti-HIV-1 potential at 100 μM ( Figure  12A). Infection dropped from a mean of 8.16%, when cells were treated with DMSO, to means of 3.78, 4.03, 2.51, 2.76 and 2.12 %, when compounds 1 to 5 were added, respectively. These differences were statistically significant for compounds 3, 4 and 5. At 50 μM, there with compounds ( Figure 12B). Compounds 1, 2 and 5 were weakly cytotoxic to THP-1 at 100 μM with 8.74, 4.15 and 5.97% of PI + cells, respectively. In total, 30-40% of cells treated with 3 and 4 were dead at all concentrations tested. Thus, although cytotoxicity could contribute to the observed antiviral activity, this experiment was performed using a single-cycle infection system, lessening the effect of cell loss on further cycles of viral replication. In conclusion, C. biflorum contains compounds with interesting anti-retroviral activity.
No alkaloids were detected in the acid organic extract from bulbs of C. biflorum, object of the present study, using either the optimized extraction method [52] or the traditional extraction with ethanol by Soxhlet. However, the organic extract obtained with the latter
No alkaloids were detected in the acid organic extract from bulbs of C. biflorum, object of the present study, using either the optimized extraction method [52] or the traditional extraction with ethanol by Soxhlet. However, the organic extract obtained with the latter method showed the presence of four homoisoflavonoids and one alkylamide. The racemate nature of compound 1 was never reported before; only the absolute configuration of its p-bromobezoyl derivative was previously determined by X-ray [53] when configurations were also confirmed by ECD comparing their ECD spectra with those previously reported. Urgineanin A was previously reported to have antiproliferative activity at submicromolar concentration against ovarian carcinoma, melanoma and non-small lung cancer cells [31].
Compound 4 is also a racemic mixture and therefore its ECD spectrum could not be interpreted (Figure 4). This result differed from that previously reported for 4 when synthesized from the (3R)-3,9-dihydroeucomnalin and wrongly reported as R enantiomer [44]. Compound 4, as well as the starting homoisoflavonoid above described for compound 1, are a racemic mixture for the presence at C-3 of a proton which could exchange by keto-enol tautomerism.
Racemic natural products are rare and could be obtained from nonenzymatic reactions [54]. However, a chiral tertiary asymmetric carbon in α position to a carbonyl group is easily subject to racemization, as in compounds 1 and 4 [55]. To the best of our knowledge no literatures are available in which compounds 1 and 4 are reported as racemates. However, Sylao et al. 1990 [30] isolated for the first time homoisoflavanoid 1 (named compound 12) along with other compounds from Scilla nervosa subsp. rigidifolia. The previously undescribed homoisoflavanoinds were also fully characterized by spectroscopic data (UV, IR, 1 H and 13 C NMR and EIMS) but any experiments to determine the absolute configuration of the chiral compounds were neither carried out nor discussed. Among the homoisoflavanoids isolated as new compounds, only for some of them, including 12 (=1), was reported the optical specific activity. Based on our experience, these results did not surprise us as the speed of inversion of the configuration of carbon 3, due to the keto-enol tautomerism, could depend on the properties of the solution in which the measurement of the optical rotational power is measured and not from the plant source. Thus, it is possible to have also a scalemic mixture that has optical activity according to the percentage of the predominant enantiomer, as reported for some optical active homoisoflavanois by Sylao et al. 1999 [30]. Scalemic mixtures of two enantiomers are reported in the literature for some different secondary metabolites isolated from several sources as i.e.: phantasmidine, an alkaloid found to be a 4:1 scalemic mixture, enriched in the (2aR,4aS,9aS) enantiomer isolated from the poison frog Epipedobates anthonyi [56]; α-pinene, 1-octen-3-ol linalool found as a scalemic mixture of 34% (R)-(+) to 66% (S)-(−), 95% (R)-(−) to 5% (S)-(+), 96% (R)-(−) to 4% (S)-(+) when isolated from the edible wild mushroom Tricholoma magnivelare [57]; a furoic acid derivative, containing a chiral center in benzylic position, was found to be a scalemic mixture with an excess of the (S) enantiomer, when obtained from the endophytic fungus Coniothyrium sp. was isolated from leaves of Quercus robur [58]; six pairs of new 6-monosubstituted dihydrobenzophenanthridine alkaloids were separated as scalemic mixtures from the aerial part of Chelidonium majus, a plant belonging to the Papaveraceae family, which is widely used in Chinese folk medicine [59].
Compounds 5, the alkylamide, was isolated for the first time, together with N-transferuloyl octopamine, N-trans-p-coumaroyl octopamine, vanillin, isoscopoletin, ethyl caffeate, ferulic acid and p-amminonenzaldehyde, from the eggplant roots [60]. Successively, 5 was also isolated together with close alkylamides from stem parts of Annona montana (Annonaceae) and showed significant inhibition of rabbit platelets aggregation induced by thrombin, arachidonic acid, collagen and PAF (platelet-activating factor) and selective cytotoxicity against the P-388 cell line [61,62]. Compound 5 was also found together with other close alkylamides and several phenolic compounds in the methanol extract of basil, lemon thyme, mint, oregano, rosemary, sage and thyme showing antioxidant and anti-inflammatory activities [63].
Alkylamides are a group of bioactive natural compounds widely distributed in plant families and characterized by broad structural variability and a plethora of important biological activities, such as immunomodulatory, antimicrobial, antiviral, larvicidal, insecticidal, diuretic, pungent, analgesic, cannabimimetic and antioxidant activities. Furthermore, they have reinforced the efficacy of antibiotics and inhibited prostaglandin biosynthesis, RNA synthesis and arachidonic acid metabolism [64]. In addition, alkylamides accumulate in rice plants as a defense against the harmful Cochliobolus miyabeanus and Xanthomonas oryzae pathogens [65].
Flavonoids were also reported to have cytotoxic and anticancer activity, although this aspect was not deeply addressed [66,67]. We found that all tested metabolites were toxic for cancer cells, in a dose-and time-dependent way, although the degree of inhibition of cell viability was cell-type specific. Remarkably, homoisoflavanoids 1, 3 and 4 were more effective on A431 and HeLa cells compared to immortalized but not transformed HaCaT, thus suggesting that cancer cells were more sensitive to homoisoflavanoid cytotoxicity. The increase of nuclear γH2AX foci upon treatment with metabolites 3 and 4 strongly suggests the occurrence of DNA damage by double-strand breaks (DSBs). However, both 3 and 4 did not increase the production of ROS indicating that a molecular mechanism different from ROS generation is responsible for the observed DNA damage induced by 3 and 4. Additional experiments are needed to precisely define the molecular mechanism. Moreover, we also detected the signal of cleaved PARP-1 (89 kDa) by immunoblot while the level of the cell cycle inhibitor p21WAF, which causes cell growth arrest preventing the induction of apoptosis, was reduced, thus confirming that the decrease of cell viability was due to cell death rather than cell cycle arrest.
Isolated compounds displayed additional interesting biological properties. As previously described but isolated for the first time in a Crinum species, 5 is a strong anti-αglucosidase and anti-α-amylase inhibitor. Recently, the efficacy of flavanoids for type 2 diabetes mellitus (T2DM) was shown in clinical therapies. T2DM is a metabolic disorder associated with the overproduction of free radicals and oxidative stress. Diabetes is increasing exponentially, and the World Health Organization estimates that by the year 2030, it could be the seventh cause of death worldwide [68,69]. The flavonoids appear to play a role in multiple processes involved in T2DM [70,71] such as the regulation of glucose metabolism, hepatic enzymes activities and a lipid profile [72]; thus, studies on nutritional flavonoids to manage diabetes and its complications are currently in progress [73].
Interestingly, the N-p-coumaroyltyramine alkylamide (5) was selectively cytotoxic against A431 and HeLa cancer cells while it protected immortalized HaCaT cells against oxidative stress induced by hydrogen peroxide. This result is highly relevant for a potential application of 5 in anticancer therapy. Furthermore, we observed the antidiabetic properties of 5 and the anti-acetylcholinesterase activity in compounds 1-4. Homoisoflavanoids are also known for their anti-acetylcholinesterase properties, a key enzyme to Alzheimer's disease development [74]. We detected anti-acetylcholinesterase activity in compound 3, a property that had not yet been reported for urgineanins A and B, to our knowledge. Finally, as we recently showed that Crinum jagus contained antiviral compounds, we measured the antiretroviral activity of isolated compounds. We report that all compounds possessed anti-retroviral potential, 3 and 4 being the most potent inhibitors.

Conclusions
Four homoisoflavanoids and one alkylamide were isolated and characterized from C. biflorum, an Amaryllidaceae plant used in African traditional medicine, collected in Senegal. Flavonoids 1, 3 and 4 showed promising anticancer properties being cytotoxic at low micromolar concentrations towards HeLa and A431 human cancer cell lines. The N-p-coumaroyltyramine (5) was selectively toxic to A431 and HeLa cancer cells, while it protected immortalized HaCaT cells against oxidative stress induced by hydrogen peroxide. Compounds 1-4 also inhibited acetylcholinesterase activity, with compound 3 being the most potent. The anti-amylase and the strong anti-glucosidase activity of compound 5 were confirmed. This study extends the chemical library of compounds that can be a potential candidate for the treatment of cancer, viral infections, diabetes and Alzheimer's disease.