Chemical Constituents of the Bark of Zanthoxylum gilletii (Rutaceae) and Their In Vitro Antiplasmodial and Molecular Docking Studies

. Te phytochemical investigations of the methanol extract of Zanthoxylum gilletii bark led to the isolation of thirteen compounds identifed as two alkaloids including one acridone 5-hydroxynoracronycine ( 1 ) and one benzo [ c ] phenanthridine decarine ( 2 ), three lignans trans - and cis -fagaramide ( 3 and 4 ) and sesamin ( 5 ), two coumarins scoparone ( 6 ) and scopoletin ( 7 ), three pentacyclic triterpenoids fridelin ( 8 ), lupeol ( 9 ) and erythrodiol-3-O-palmitate ( 10 ), one phenolic compound vanillic acid ( 11 ) as well as two common steroids stigmasterol ( 12 ), and its derivative stigmasterol-3-O-β -D-glucopyranoside ( 13 ). Te structures of all the isolated compounds were elucidated by means of their spectroscopic and spectrometric data (1D, 2D-NMR, MS) as well as the comparison of these data with those reported in the literature. Except for compounds 9 and 11 – 13 , all the other isolated compounds are reported for the frst time from Z. gilletii but have been already obtained from other Zanthoxylum species and in the Rutaceae family. Compounds 1 , 3 – 5 , and 9 were tested in vitro for their antiplasmodial potencies against Plasmodium falciparum 3D7, and the results revealed that all the tested compounds displayed an inhibition between 51.89% and 54.69% while only the mixture of 3 + 4 gave an IC 50 lower than 10 000nM (IC 50 � 1333nM). Furthermore, all the compounds have been evaluated in silico for their ability to inhibit the Plasmodium falciparum dihydroorotate dehydrogenase 5TBO. Sesamin ( 5 ) showed the greatest afnity to the antiplasmodium receptor than artemether ® and chloroquine ® . Further recorded data from their ADMET study, as well as their chemotaxonomy, are also discussed herein. Te present study provides further information to enrich the chemistry of Z. gilletii and its qualifcation as an important source for good candidates in new antiplasmodial drug development.


Introduction
Malaria is a parasitic disease with a great incidence in the public health sector in many countries in tropical areas of the world [1]. A recent report by the World Health Organization (WHO) indicated that 228 million cases of malaria have been recorded in the world with 405 thousand deaths including almost 94% occurring in Africa. More specifcally, ffteen African countries accounted for 80% of global malaria deaths while 2,974,819 confrmed cases and an estimation of 14,841 deaths have been reported in Cameroon during the year 2020 [2,3]. Te disease is caused by the parasite Plasmodium transmitted through the bites of the mosquitoes Anopheles [1,4]. It is well reported that medicinal plants represent an important source of bioactive compounds that can be considered as good candidates in the development of new potent drugs. For instance, the well-known antimalarial drug artemisinin was initially obtained from the plant Artemisia annua while quinine was isolated from Cinchona ofcinalis [5]. Unfortunately, the observed rise of resistance of P. falciparum to the available antimalarial drugs keeps actual the search for new afordable and potent compounds that can be developed as drugs. Zanthoxylum gilletii (Rutaceae) also called Z. macrophylla, Fagara tessmanii, or F. macrophylla is a tree encountered in the tropical rainforest that can reach 10 to 35 m high [6,7]. Te plant is one of the 2000 species that constituted the genus Zanthoxylum and its synonyms in the genus Fagara which are widely used in African folk medicine for the treatment of several diseases [8]. Te stem bark of Z. gilletii is used in Cameroon and Madagascar for the treatment of microbial infection, cancer, infammation, hypertension, and related disorders while its bark is used in Kenya and Ivory Coast against malaria [8]. Te use of Z. gilletii in the management of malaria has been further supported by the good antimalarial activity of its ethanolic extract against a panel of Plasmodium falciparum strains with an IC 50 value of 5 μg/mL [9]. As a continuity of our program in searching for antiplasmodial agents from Cameroonian medicinal plants [1,10], we carried out phytochemical investigations of the bark of Z. gilletii. In addition, the in silico study of the isolated compounds as potential inhibitors of the 5TBO receptor (Plasmodium falciparum dihydroorotate dehydrogenase) as well as their ADMET evaluation and chemotaxonomy have been done, and the results are herein discussed. Tis study provides additional compounds to enrich the chemistry of Z. gilletii, its chemotaxonomic classifcation in the family Rutaceae as well as it further supports its use in folk medicine by the local population for the management of malaria and related symptoms.

General Instrumentation.
EI-MS were recorded on a Finnigan MAT 95 spectrometer (70 eV) (Termo Fisher Scientifc, Darmstadt, Germany) with perfuorokerosene as a reference substance for EI-HR-MS. Te spectrometer operated in positive and negative modes (m/z: 50-1500, with a scan rate of 1.00 Hz) with automatic gain control to provide high-accuracy mass measurements within 1 ppm deviation using Na formate as a calibrant. Te following parameters were used for experiments: spray voltage of 4.5 kV and the capillary temperature of 200°C. Nitrogen was used as sheath gas (4 l/min). Te 1 H-and 13 C-NMR spectra were recorded at 500 MHz and 125 MHz, respectively, on Bruker AMX 500 NMR spectrometers (Bruker, Rheinstetten, Germany). Chemical shifts are reported in δ (ppm) using TMS as an internal standard, and coupling constants (J) were measured in Hz. Column chromatography was carried out on silica gel (70-230 mesh (Merck, Darmstadt, Germany). Tin-layer chromatography (TLC) was performed on Merck precoated silica gel 60 F 254 aluminium foil (Merck, Darmstadt, Germany), and spots were detected using ceric sulphate spray reagent. All reagents used were of analytical grade.

Extraction and Isolation.
Te yellowish bark of Z. gilletii (3.8 kg) has been dried and grounded to obtain a powder that was twice extracted with methanol at room temperature for 48 h, each, to aford 148.3 g of crude extract after removing solvent. After keeping a small amount for biological tests and further chemical analyses, 130 g of crude extract was dissolved in water and successively partitioned with ethyl acetate (EtOAc) and n-butanol to give two main fractions A (78.4 g) and B (46.8 g). Te ethyl acetate soluble fraction A has been further chromatographed over a silica gel column (230-400 mesh, 5.0 × 75.0 cm) eluting with increasing polarity of ethyl acetate in n-hexane (n-Hex) followed by the gradient of methanol in ethyl acetate. A total of 563 subfractions (ca. 200 mL each) were collected and combined into 6 series A1-A6 based on their TLC profles.
Finally, the series A4 (10.3 g, n-Hex/EtOAc 1 : 1) and A5 (11.4 g, EtOAc) were mixed and chromatographed using a gradient of methanol in dichloromethane to obtain compounds 7 (2.8 mg) and 13 (28.7 mg), while the last series A6 (18.1 g, AE/MeOH 5%) has been mixed to the n-BuOH soluble main fraction B and subjected to column chromatography using the increasing polarity of methanol in EtOAc to aford compound 13 (3.8 mg), and some precipitates are insoluble in the organic solvent.

Plasmodium falciparum Culture and Growth Inhibition
Assay. Plasmodium falciparum 3D7 (chloroquine-sensitive) strain was obtained from the Biodefense and Emerging Infections (BEI) Research Resources (Manassas, VA) and maintained using a modifed Trager and Jensen method. Te biological assay has been done following a modifed method of Yemback et al. [11]. Briefy, parasites were cultured in fresh O + human red blood cells at 3% (v/v) hematocrit in RPMI 1640 culture media containing GlutaMAX and NaHCO 3 (Gibco, UK) and supplemented with 25 mM HEPES (Gibco, UK), 1X hypoxanthine (Gibco, USA), 20 µg/ mL gentamicin (Gibco, China), and 0.5% Albumax II (Gibco, USA). When needed, parasites were synchronized at the ring stage by sorbitol treatment and cultured through one cycle before treatment.
Stock solutions (10 mM) were prepared in incomplete RPMI 1640 and mixed with parasite cultures (1% parasitemia and 1.5% hematocrit) in 96-well plates to a fnal drug concentration of 10 µM for hit identifcation studies, or 10-0.078 µM for activity confrmation assays. Te fnal dimethyl sulfoxide (DMSO) concentration per 100 μL culture per well was 0.1%. Chloroquine and artemisinin at a range of 1-0.0078 µM each were used as negative growth control (positive test control), while the solvent-treated culture (0.1% DMSO) was used as positive growth control (negative test control). Following 72 h incubation at 37°C in a 5% CO 2 incubator, parasite growth was assessed by an SYBR green I-based DNA quantifcation assay. In brief, a 3x concentrated SYBR Green lysis bufer was added to each plate well containing parasitized erythrocytes and kept in the dark for about 30 minutes. Fluorescence was measured using a Fluoroskan Ascent multiwell plate reader with excitation and emission wavelengths at 485 and 538 nm, respectively. Mean half-maximal inhibitory concentrations (IC 50 values) were derived by plotting percent growth against log drug concentration and ftting the response data to a variable slope sigmoidal curve function using GraphPad Prism v8.0.

Computational Tools.
All in silico studies applied to this work were achieved using the following computational tool: (a) ACDChemSketch ® version 2.0, year 2020 used for modeling, frst-hand, chemical structures to be fed for docking screening [12] (b) Autodock Vina ® tool version 1.2.0, year 2010 used for the virtual screening procedure [13] (c) BIOVIA ® Discovery Studio version 2021 used for the determination of binding pockets and for visualizing/analyzing the docked results [14] (d) Absorption Distribution Metabolism Excretion and Toxicity of Structure Activity-Relationship (ADMETSAR): a web server used to evaluate the ADMET properties, biological activity, and druglikeness [15][16][17][18]

Molecular Docking
(1) Ligand Preparations. In a quest to predict the antiplasmodial activity of the plant samples, compounds present in Z. gilletii were modeled using the ACDChem Chem-Sketch ® version 2.0, year 2020 [12] computational tool and fed into Autodock vina ® [13] for binding efciency screening.
(2) Receptor preparations. Te antiplasmodium receptor of PDB CODE, 5TBO, was used as the target receptor for this study. Te receptor was obtained from the (RCSB) database (https://www.rcsb.org/pdb). Te resolution of receptor 5TBO (2.15Å) approximately lies about the 2.0Å recommended resolution for a receptor of good quality [19]. Water molecules around the crystal structure and other associated inhibitors were removed from the downloaded protease to avoid undesired interferences and unwanted molecular interactions. Discovery Studio version 2021 [14]. Te screening was guided using Chloroquine ® and Artemether ® , prominently used antiplasmodium therapeutics, as standards. Auto Dock Tools version 4.0 (ADT) was used to generate the autodock fle (.pdbqt fle) in preparation for docking. Other adjustments performed on both crystal structures include removing any extra water molecules, adding polar hydrogens, merging no-polar hydrogens, and adding Kollman charges. Tese adjustments are targeted at improving the afnity of the binding site. In modern computer-assisted drug design (CADD), the study of a potent drug is not complete without the activity study of the drug which is deducible from the compound's inhibition constant. A potent compound is expected to have inhibition constant values ranging between 0.01 and 1 μM for a potent drug; therefore, this parameter alone is enough to screen compounds in the drug discovery journey [19].

3.2.
In Vitro Antiplasmodial Activity. Among the isolated compounds, those in large amounts (compounds 1, 3 + 4, 5 and 9) were submitted for in vitro antiplasmodial activity against the chloroquine-sensitive P. falciparum 3D7 (Pf 3D7) using chloroquine (IC 50 � 24.74 nM) and artemisinin (IC 50 � 17.76 nM) as reference drugs. All the compounds displayed good inhibition rates measured as 1 (54.69%), 3 + 4 (53.48%), 5 (51.89%), and 9 (52.64%). However, the most potent compound was the mixture 3 + 4 with an IC 50   value of 1333 nM while compounds 1, 5, and 9 were less or not active with IC 50 values greater than 10,000 nM (Table 2). Te inhibition level greater than 50% for all the compounds showed that the Z. gilletii extract is an important source of antiplasmodial compounds supporting the use of the plant in traditional medicine for the treatment of malaria and deserves further investigations in checking of synergistic efects of those compounds as well as the characterization of further constituents that can help in the development of new potent antimalarial drugs.

Molecular Docking.
Te binding afnities of all the thirteen compounds isolated from Z. gilletii as well as the selected standards (Chloroquine ® and Artemether ® ) are given in Table 3 and Tis can also infer that the plant Z. gilletii has a high potency as an antiplasmodial agent. It can also be mentioned that since both standards used for this study (Chloroquine ® and Artemether ® ) resulted in a maximum number of two conventional hydrogen bonds (Figure 8) while most selected compounds exhibited more, and with shorter bond length, it is understandable why the identifed compounds have greater binding afnity to the antiplasmodium receptor than the standards.

ADMET Studies.
Te compounds displaying a good afnity (binding energy) to the 5TBO receptor binding site were further analysed for their ADMET properties, and the prediction results are presented in Table 1. Te human intestinal absorption (HIA) of an oral drug is the essential prerequisite for its apparent efcacy. Furthermore, the close relationship between oral bioavailability and intestinal absorption has also been proven, and HIA can be seen as an alternative indicator for oral bioavailability to some extent. A drug molecule with an absorbance of less than 30% is poorly absorbed [20]. Te results from Table 1 show that compounds 1-3, 5, 8-9, 12-13 as well as both standard drugs (chloroquine and artemether) are strongly absorbed by the intestine.
Water solubility is one important condition that a drug must satisfy before the in vitro drug optimization process [20]. In this regard, the results show that all the identifed compounds except compound 12 are water-soluble with −3.615, −4.305, and −4. 793 for compounds 1, 3, and 13, respectively, which are more soluble than the selected standards, chloroquine (−3.902) and artemether (−4.677).
Te human either-a-go-go related inhibition (hERG) is a measure of electrical pulse transmission in the human heart (iodine-channel blocking). A potential drug should not block the activity of the hERG inhibitors [34]. Table 1 shows that all compounds present themselves as inactive against blocking the heart functions, as the standards. Carcinogenicity results also show that compounds 8, 9, 12, and 13 are noncarcinogenic to mammals. However, compounds 1, 2, 3, and 5 are seen to be toxic.
Brain-blood barrier (BBB) prediction test suggests that compounds 1, 2, 5, 9, 12, and 13 do not have the ability to cross the BBB; hence, their modes of action do not involve neurological engineering, such as appetite reduction as in chloroquine. It also suggests that the six (6) compounds will pose little or no side efects on administration as compared to chloroquine. However, compounds 3 and 8 have a tendency to cross the BBB.
A good drug must also be able to stay long (activity) enough in the body before excretion [35]. Te half-life of a drug is the description of how long a drug will stay active in the body before excretion. Table 1 indicates that all identifed compounds can exhibit a better excretion than the standard drugs.
In addition, Table 1 also describes the drug likeliness based on Lipinski's rules of fve (5) criteria [36]. Te rule  Journal of Chemistry 5 states that for a compound that must stand as a potential drug, its structural properties must frst satisfy the following rules: molecular weight ≤500; MiLOGP ≤4.15; number of oxygen ≤10; number of hydrogen ≤5; number of hydroxide ≤5.
Results show that all identifed compounds satisfy this drug condition and hence are more likely to be applied as a drug.
Except compound 10 which is obtained here for the frst time from the genus Zanthoxylum and in general from the family Rutaceae, all the other compounds have been previously reported from the genus Zanthoxylum or from the genus Fagara which contains several species which are synonyms of those classifed in the genus Zanthoxylum. For instance, trans-fagaramide (3), scopoletin (7), and vanillic acid (11) have been previously reported from the bark of Fagaramacrophylla and Fagara tessmannii [23,37]. Tese two species are both synonyms of Z. gilletii, the plant we have investigated. Tis result partially supports the taxonomy of our plant and its similarity with the two Fagara species based on their chemical constituents. Further evidence has been given by the identifcation of the other metabolites: the acridone alkaloids have been mainly isolated from the

Journal of Chemistry
Rutaceae plants and more especially from the plants from the genus Citrus. Unsurprisingly, the acridone 5-hydroxynoracronycine (1) is one of the major compounds along with sesamin (5) which were already reported from the plant species Z. poggei and Z. dinklagei [21,26], respectively, as well as for the frst time to the best of our knowledge from Z. gilletii. Another lignan named cis-fagaramide (4) has been also isolated during our investigations but has been already identifed by Cheng et al. in [16] during their works on Z. schinifolium. Te alkaloid decarine (2) is obtained for the frst time from Z. gilletii while it has been found in Z. madagascariense [22]. Furthermore, the coumarin scopoletin (6) is widely encountered in many plant species and has been detected and characterized from the species Z. rhetsa, Z. schinifolium, and Z. leprieurii [27,38,39]. Te two triterpenoids fridelin (8) and lupeol (9) are also mainly distributed in several plants and within the family Rutaceae, and they have been isolated from Z. schinofolium, Z. tessmannii, or Citrus aurantium [24,29,40]. However, the third triterpene identifed as erythrodiol-3-O-palmitate (10) has not been previously obtained from the family Rutaceae as indicated earlier. Compound 10 is a derivative of β-amyrin belonging to the oleanane-type triterpenoids [41]. Te literature survey indicates that some congeners of compound 10 including β-amyrin or β-amyrin acetate have already been reported from the Zanthoxylum genus. Indeed, β-amyrin was obtained in Z. simulans and Z. nitidum [42,43], while β-amyrin acetate was reported from the species Z. schinifolium [24]. Tis evidence strongly supported the possibility of obtention of erythrodiol-3-O-palmitate (10) which is very structurally close to β-amyrin acetate and difers from it by just the length of the fatty acid attached to C-3 as well as the oxidation of the methyl CH 3 -28 to a primary alcohol function CH 2 OH [41]. Erythrodiol-3-O-palmitate (10) has been previously obtained from Trilepsium madagascariense (Moraceae) [32] and is reported for the frst time from Z. gilletii and the family Rutaceae. Finally, the steroids stigmasterol (12) and stigmasterol-3-Oβ-D-glucopyranoside (13) are usually obtained from several plant materials. Tus, all the chemical compounds isolated from Z. gilletii provide clear insights supporting the taxonomy of the plant and enrich the knowledge of its chemistry.

Concluding Remarks
Tirteen compounds (1-13) have been isolated and characterized from the phytochemical investigations of the methanol extract of the bark from Z. gilletii, a Cameroonian medicinal plant. Te compounds have been sorted into six main classes of compounds including two alkaloids (1 and 2), three lignans (3)(4)(5), two coumarins (6 and 7), three pentacyclic triterpenoids (8-10), one phenolic compound (11) as well as two steroids (12 and 13). Te chemotaxonomic signifcance of the isolated compounds strengthened the classifcation of the plant in the Rutaceae family and provided additional compounds to enrich the chemistry of the species Z. gilletii (Rutaceae). More interestingly, all the compounds tested in vitro showed good inhibition rates of Pf 3D7 with values greater than 50% while the molecular docking of the isolated compounds showed that a total of eleven compounds displayed a good afnity with the 5TBO receptor binding site compared to the standard drugs chloroquine and artemether. Likewise, the ADMET study indicated that isolated compounds exhibited good and favourable parameters that qualify them as good candidates for drug development. Te present study, in addition to supporting the use of the plant in traditional medicine for the treatment of malaria, gives signifcant insights for further in vitro and in vivo pharmacological investigations in the discovery of new antimalarial drugs.

Data Availability
Te spectroscopic ( 1 H and 13 C-NMR) and spectrometric (EI-MS) data (Table 1S) of all the isolated compounds and their molecular docking data (Table 2S) used to support the fndings of this study are included within the supplementary information fle.

Conflicts of Interest
Te authors declare that they have no conficts of interest. Table 1S shows the mass and NMR data of isolated compounds.