Phytochemical analysis of Acacia nilotica and evaluation of its antileishmanial potential

: Acacia nilotica is an important medicinal plant, found in Africa, the Middle East, and the Indian subcontinent. Every part of the plant possesses a wide array of biologically-active and therapeutically important compounds and has been used in the traditional system of medicine. We reported the antileishmanial activity of Acacia nilotica ( A. nilotica ) bark methanolic extract through in vitro assays and dissected the mechanism of its action through in silico studies. Bark methanolic extract exhibited anti-promastigote and anti-amastigote potency with IC 50 value of 19.6 + 0.9037 µg/ml and 77.52 + 5.167 µg/ml respectively in time and dose-dependent manner. It showed very low cytotoxicity having a CC 50 value of 432.7 + 7.71 µg/ml on the human-macrophage cell line, THP-1. The major constituents identified by GC-MS analysis are 13-docosenoic acid (34.06%), lupeol (20.15 %), 9,12-octadecadienoic acid (9.92 %), and 6-octadecanoic acid (8.43 %) bind effectively with the potential drug-targets of Leishmania donovani ( L. donovani ) including sterol 24-c-methyltransferase (SMT), trypanothione reductase (TR), pteridine reductase (PTR1) and adenine phosphoribosyltransferase (APRT); suggest the possible mechanism of its antileishmanial action. The highest affinity with all these targets was shown by lupeol. The pharmacokinetic studies, predicted bioactivity scores, and acute toxicity studies of major extract constituents support safe antileishmanial drug candidates. This study proved the antileishmanial potential of bark-methanolic extract A. nilotica and its mechanism of action through the inhibition of potential drug targets of L . donovani.

We also tried to dissect the mechanism of its antileishmanial action through different in silico approaches. SMT, TR, PTR1, and APRT are prerequisite enzymes for survival, pathogenicity, and transmission of L. donovani. Therefore, we selected these potential drug-targets for molecular docking study of major constituents of bark-extract identified by GC-MS, with these mentioned essential enzymes of Leishmania.

Materials and method
2.1 Chemicals: M199 media, Roswell park memorial institute (RPMI) 1640 media, penicillinstreptomycin antibiotic cocktail, fetal bovine serum (FBS) were purchased from Gibco. HEPES, sodium bicarbonate, and paraformaldehyde were purchased from Sigma Aldrich. Miltefosine, MTT assay reagents, DMSO, and different solvents were procured from Merck. Propidium iodide and Annexin V apoptosis kit were procured from Thermo scientific. All the other chemicals and reagents were purchased from Sigma Aldrich or Merck unless stated otherwise.

Parasites and cell culture:
The infective strain of L. donovani (MHOM/IN/83/AG83) was obtained from Dr. Rentala Madhubala (School of life science JNU, New Delhi; India). THP-1, a human monocytic cell line was procured from the Cell Repository of National Centre for Cell Science, Pune, India. It was further maintained in M199 media. Human monocytic cell line, THP-1 was maintained in RPMI 1640 media supplemented with 10 % FBS and 1 % penicillin-streptomycin antibiotic medium in a humidified environment at 5 % CO2 and 37 0 C temperature. THP-1 monocytic cell was differentiated to macrophages by using phorbol myristate acetate (PMA) at a concentration of 20 ng/ml.

Extract preparation and antileishmanial activity:
A. nilotica was collected from natural habitats. Bark identification was done at the National Institute of Science Communication and Information Resources (NISCAIR), New Delhi, India.
The selected plant material was washed and air-dried in shade at room temperature. The powdered plant materials were soaked in methanol and placed on the rotary shaker at room temperature for 24 h. The extract was filtered and concentrated using a rotatory evaporator under vaccum at 35 0 C. The dried plant extract was stored at -20 0 C until used for bioassay. To evaluate the anti-promastigote potential of A nilotica, stationary phase (2 x 10 6 cells/ml) promastigotes were incubated with plant extract for 48 h followed by fixing using 1 % paraformaldehyde and counting through hemocytometer at 22 0 C. Miltefosine a known antileishmanial drug used as positive control. Percent viability was determined using the formula: 50 % inhibitory concentration (IC50) at which parasite growth was reduced by 50 % was assessed by GraphPad Prism 7.00, nonlinear regression curve fit.

Cytotoxicity assessment and anti-amastigote evaluation of extract:
The cytotoxicity of A. nilotica on THP-1 differentiated macrophages was assessed by MTT [3-(4,5 dimethyl-thiazol-2-yl)-2,5-diphenyl tetrazolium bromide]. Briefly, 2 x 10 6 THP-1 monocytes were seeded in 96 well tissue culture plate (200 μl/well) in RPMI1640 complete media for 24 h in a CO2 incubator at 37 0 C and 5% CO2. After treatment to THP-1 differentiated macrophages, freshly prepared 5 mg/ml MTT was added (20 μl/well) with 50 μl of blank media % Viability and further incubated for 2 to 3 h in a CO2 incubator. Precipitated formazan was dissolved in dimethyl sulfoxide (DMSO) absorbance was recorded at 570 nm in an ELISA plate reader and percent viability was calculated. To determine the effect of A. nilotica on the parasite burden of the host macrophages, 0.5 x 10 6 THP-1 cells were seeded on the coverslip, placed in the six-well plates in a CO2 incubator at 37 0 C. THP-1 macrophages were plated, infected with L. donovani at the ratio of 1:10 (macrophages to Leishmania) for 48 h then cells were fixed with chilled methanol and parasite counting was performed under the microscope after Giemsa staining.
From the different focus, 100 macrophages were counted to determine the parasite burden of the macrophages. Parasite burden in the infection control was considered 100 %, with respect to parasite load in treated samples.

GC-MS analysis of extract:
GC-MS analysis was performed to identify the secondary metabolites that may be responsible for the antileishmanial efficacy of A nilotica. Bark was crushed, powdered, and extracted in methanol and then analyzed on Shimadzu QP2010 armed with a DB-5MS column. The mass spectrums of the sample were produced in an electron impact ionization mode of 70 eV and the phytochemicals were identified after correlation of the recorded mass spectrum with the reference library WILEY8.LIB and NIST14.LIB supplied with the software of the GC-MS system.

Molecular docking studies:
To begin with structure-based virtual screening and docking, we used various bioinformatics tools, such as PyRx 25 , AutoDock Vina 26 , PyMOL 27 , and BIOVIA Discovery Studio. 28 The online resources used in the retrieval, analysis, and evaluation of the data are the PubChem database and RCSB Protein Data Bank (PDB). 29 The target proteins of L. donovani and the phytochemical compounds were uploaded into the virtual screening program PyRx. The target protein was changed into a macromolecule, which converted the atomic coordinates into pdbqt format. Molecular docking was performed by selecting the grid box around the crystal structures and the rest of the parameters were left as default. AutoDock Vina was used to predicting the binding mode and the best binding affinity of the phytochemicals. The algorithm used by AutoDock Vina is a hybrid scoring function that is inspired by X-score, which accounts for hydrogen bonding, hydrophobic effect, van der Waals forces, and deformation penalty. Besides, for computing, the binding energy AutoDock Vina combines both the conformational preferences of receptor-ligand complex and experimental affinity measurements. The results of molecular docking were screened for binding affinity and then all possible docked conformations were generated for different constituents. After analyzing with PyMOL and Discovery Studio, only those conformations were selected which specifically interact with the active-site residues of L. donovani targeted proteins. Discovery Studio was used to analyze detailed interactions and their types including hydrogen bonds, alkyl, pi-alkyl, halogen, and the van der Waals interactions formed between different constituents and the target proteins. The most favorable binding poses of the rutin were analyzed by choosing the lowest free energy of binding (ΔG) and the lowest inhibition constant (Ki) which is calculated using the following formula: where ΔG is binding affinity (kcal/mol), R (gas constant) is 1.98 calK −1 mol −1 , and T (room temperature) is 298.15 Kelvin.

Sequence analysis, template identification, homology modeling, and receptor and ligand preparation
The protein sequences of trypanothione reductase (XP_003858222.1) and sterol 24-cmethyltransferase (XP_003865366.1) from L. donovani were retrieved from NCBI. The blastP 30 was performed against Protein Data Bank for the identification of similar templates. The alignment of the query sequences and template sequences was performed using CLUSTAL Ω. 31 The crystal structure of trypanothione reductase from L. infantum 2.95 Å resolution (PDB id: 2JK6_A) and X-ray diffracted crystal structure 1.34 Å resolution (PDB id: 5WP4_A) were used as template structures to model the 3D structures of trypanothione reductase and sterol 24-cmethyltransferase, respectively. PDB was used to retrieve the template structure. Homology modelling was carried out using Modeller 9.24 32 and PyMol was used for the visualization of the 3D structures. The energy minimization was performed using Discovery studio. The PROCHECK program, Ramachandran plots were also used for the assessment of the model. 33 Crystal structures of the Adenine phosphoribosyltransferase and Pteridine reductase proteins were downloaded from PDB [IDs: 1QB7 (APRT) and 2XOX (PTR1)]. The PDB files used for the docking-based virtual screening study were processed by removing water molecules and adding hydrogen atoms. The proteins were finally prepared by Discovery Studio keeping all the parameters at default. The identification of the critical residues of the binding pockets was taken from the native binding pockets of the available crystal structure of proteins, various submitted literature, from their homologous template proteins, and investigation in the mechanism of inhibition. The 3D structure of 9,12-Octadecadienoic acid, 6-Octadecenoic acid, 13-Docosenoic acid, and Lupeol was retrieved from the PubChem database in SDF format. The atomic coordinates of all the ligands were changed to pdbqt set-up using Open Babel GUI, an open-source chemical toolbox for the interconversion of chemical structures. 34 Universal Force Field (uff) was used for the energy minimization. 35

Pharmacokinetics studies
The selected ligands were evaluated for their pharmacological profiles by analyzing for Miltefosine treatment also exhibited similar morphological changes as extract showed at higher doses ( Figure 1C).

Growth reversibility assay after extract treatment:
A. nilotica, treated and untreated parasites were washed with PBS after 7 days, and old media was removed and supplemented with fresh media. The samples were further incubated at 22 0 C for the next 72 h to study the growth reversibility of parasites. Parasites treated with higher doses do not revert though parasites in flasks of lower dose plant-extract treatment show slower growth reversion ( Figure 1D). Suppression of growth reversion was observed significant (P < 0.001) at 250 µg/ml of A. nilotica in comparison to the untreated sample ( Figure 1D). Miltefosine was taken as a positive control (Figure 2A). Cell cytotoxicity (CC50) of A. nilotica methanolic extract was evaluated along with miltefosine as a positive control on THP-1 differentiated macrophages to study its safe dose. THP-1 differentiated macrophages were incubated with different concentrations of extract/miltefosine (0 to 1000 µg/ml) and the cell viability was assessed using MTT assay. It was observed that A. nilotica has the least cytotoxic effect on the viability and morphology of the macrophages with a CC50 value of 432.7 + 7.71 µg/ml while miltefosine showed higher toxicity with a CC50 value of 8.219 + 0.6337 µg/ml ( Figure 2B). A significant reduction in intra-macrophagic parasite count was observed in the micrographs of Giemsa stained infected and extract treated macrophages ( Figure 2C).

TLC-bioautography identification and GC-MS analysis of A. nilotica bark methanolic extract:
Plant secondary metabolites present in A. nilotica bark methanolic extract fractions that may have been responsible for the observed antileishmanial effects were identified through TLCbioautography and GC-MS analysis. The total constituents found were 25 (Table 1)

Molecular docking of A. nilotica methanolic extract of major constituents with potential drug-targets of L. donovani
The TR and SMT enzymes were modelled using Modeller 9.24 and the energy minimization was carried out by BIOVIA Discovery studio. The three-dimensional cartoon representation of TR and SMT enzymes is shown in Figure S1A,2A. The models were selected by analyzing their stereochemical quality using the PROCHECK program. The generated models of TR and SMT show a good quality structure having 99.8 % and 99% residues in the allowed regions of the Ramachandran plot respectively ( Figure S1B,2B). PDBsum tool was used to analyze and found that the 3D structure of the enzyme is composed of mixed α-helices and β-strands (α+β) secondary structures. 33 The structural topology of TR and SMT showed 5 sheets, 23 (Table 2). It shows favorable interactions with SMT through two pi-alkyl bonds with Arg347 and Lys351, TR via a pi-alkyl bond with Tyr198, PTR1 by two pi-alkyl bonds with Val83 and Arg88, and APRT through a hydrogen bond with Thr151. (Figure 3-6B).  and Ar82. (Figure 3-6H).

Pharmacokinetics studies of A. nilotica bark-methanolic extract constituents:
The pharmacological studies were done for the selected ligands against Adenine  Table 3.
The bioactivity prediction of the major constituents of A. nilotica bark-methanolic extract was analyzed through molinspiration. The activity was calculated against G-protein coupled receptorligand, ion channel modulator, a kinase inhibitor, nuclear receptor ligand, protease inhibitor, and enzyme inhibitor. 40 The interpreted values for bioactivity were as: active (bioactivity score ≥ 0), moderately active (bioactivity score: between −5.0 to 0.0), and inactive (bioactivity score ≤ −5.0). 41 Lupeol, 9,12-Octadecadienoic acid, 6-Octadecenoic acid, and 13-Docosenoic acid were evaluated as active enzyme inhibitors with values 0.52, 0.23, 0.12, and 0.10, respectively. Lupeol and 9,12-Octadecadienoic acid were evaluated as active protease inhibitors as well as ion channel modulators. (Table 4) The principal aim of predicting acute toxicity is to evaluate undesirable side effects of a compound after single or multiple exposures to an organism via a known administration route (oral, inhalation, subcutaneous, intravenous, or intraperitoneal). GUSAR was used to determine the acute toxicity of the successfully docked compounds. The parameters used by GUSAR to probe compounds based on the prediction of activity spectra for substances algorithm and quantitative neighborhoods of atoms descriptors. The obtained results were compared with SYMYX MDL Toxicity Database to further categorize them based on the Organisation for Economic Co-operation and Development (OECD) chemical classification manual. 36 The criteria used for these compounds to elicit toxicity based upon the administration route when the compound dose is more than 7000 mg/kg for an intravenous route, more than 500,000 mg/kg in case of the oral route, and more than 20,000 mg/kg for intraperitoneal route and subcutaneous database as shown in Table 5. As per the OECD chemical classification 9,12-Octadecadienoic acid, 6-Octadecenoic acid, 13-Docosenoic acid found to be non-toxic and Lupeol is a Class 5 chemical.

Discussion:
Plant extracts have promising medicinal properties and are extensively used in the traditional system of medicine due to the presence of many active and leading medicinal. 42  conjugated pterins such as reduced biopterin to dihydrobiopterin. 52 APRT plays a vital role in purine metabolism by converting 6-aminopurines into 6-oxypurines. 53 Molecular docking results prove that lupeol and 9,12-Octadecadienoic acid possesses higher binding affinity with SMT, TR, PTR1, and APRT as shown in Table 2. Pharmacological studies of these selected inhibitors for the Lipinski rule of 5 indicated violation of only one Lipinski parameter, as shown in Table 3.
The pharmacokinetic properties and acute toxicity of lupeol; 9,12-Octadecadienoic acid;6-Octadecenoic acid and 13-Docosenoic acid have shown relatively low toxicity profile, which means require high doses to evoke a toxic response. The majority of the compounds are nontoxic chemicals whereas lupeol is a Class 5 chemical with very low toxic effects. 54 The pharmacokinetic attributes are in favor of these compounds to be exploited as promising antileishmanial drug candidates. It had already been reported that lupeol activates the PI3K/Akt pathway which triggers mechanisms responsible for influencing various cell types, including keratinocytes, stimulated cytotoxicity in fibroblasts, and the regulation of various diseases. 55 The lupeol treatment had shown a high imbalance between Th1/Th2 cytokines production and initiation of pro-inflammatory cytokine response as well as the generation of NO in L. donovani infected macrophages. 56 It further supports the antileishmanial activity of A. nilotica extract.
Thus, in vitro, molecular docking, pharmacokinetics studies, bioactivity scores, and acute toxicity studies support possible mechanisms of antileishmanial activity of the extract through inhibition of key Leishmania enzymes.     Tables legends: