In-vitro and in-silico antibacterial activity of Azadirachta indica (Neem), methanolic extract, and identification of Beta.d-Mannofuranoside as a promising antibacterial agent

Background Antimicrobial resistance became the leading cause of death globally, resulting in an urgent need for the discovery of new, safe, and efficient antibacterial agents. Compounds derived from plants can provide an essential source of new types of antibiotics. A. indica (neem) plant is rich in antimicrobial phytoconstituents. Here, we used the sensitive and reliable gas chromatography-mass spectrometry (GC–MS) approach, for the quantitative and quantitative determination of bioactive constituents in methanolic extract of neem leaves grown in Sudan. Subsequently, antibacterial activity, pharmacokinetic and toxicological properties were utilized using in silico tools. Results The methanolic extract of neem leaves was found to have antibacterial activity against all pathogenic and reference strains. The lowest concentration reported with bacterial activity was 3.125%, which showed zones of inhibition of more than 10 mm on P. aeruginosa, K. pneumoniae, Citrobacter spp., and E. coli, and 8 mm on Proteus spp., E. faecalis, S. epidermidis, and the pathogenic S. aureus. GC–MS analysis revealed the presence of 30 chemical compounds, including fatty acids (11), hydrocarbons (9), pyridine derivatives (2), aldehydes (2), phenol group (1), aromatic substances (1), coumarins (1), and monoterpenes (1). In silico and in vitro tools revealed that.beta.d-Mannofuranoside, O-geranyl was the most active compound on different bacterial proteins. It showed the best docking energy (-8 kcal/mol) and best stability with different bacterial essential proteins during molecular dynamic (MD) simulation. It also had a good minimum inhibitory concentration (MIC) (32 μg/ml and 64 μg/ml) against S. aureus (ATCC 25,923) and E. coli (ATCC 25,922) respectively. Conclusion The methanolic extract of A. indica leaves possessed strong antibacterial activity against different types of bacteria. Beta.d-Mannofuranoside, O-geranyl was the most active compound and it passed 5 rules of drug-likeness properties. It could therefore be further processed for animal testing and clinical trials for its possible use as an antibacterial agent with commercial values. Supplementary Information The online version contains supplementary material available at 10.1186/s12870-022-03650-5.

Altayb et al. BMC Plant Biology (2022) 22:262 Background Medicinal plants are known to have a wide range of bioactive compounds that have antimicrobial, antifungal, anticancer, anti-inflammatory, and antioxidant activities [1][2][3]. Many researchers have documented the potent activity of plants' bioactive compounds on drug-sensitive and resistant bacteria [4][5][6]. Although plants contain a very large number of bioactive compounds, few have been discovered [1]. The development of extraction methods and the use of molecular spectroscopic techniques such as GC-MS and fourier-transform infrared (FTIR) has led to the discovery and characterization of new plant bioactive compounds [1,[7][8][9]. Recently, in silico tools have emerged as promising, time and costsaving approaches for drug discovery [10].
Azadirachta indica (A. indica) is one of the Meliaceae family known as neem. It has been used in traditional medicine since ancient times to treat a range of human diseases [11]. The leaves, seeds, and roots of neem contain antibacterial and antifungal agents [12,13]. This biological activity of neem stems from many bioactive compounds that are structurally and chemically diverse, with more than 140 compounds found in different parts of the plant [14]. Several types of biological compounds are extracted from neem, including ketones, carotenoids, flavonoids, steroids, and phenolic compounds [15]. The antibacterial activity of A. indica leaves extract has been documented on different bacterial species including E. coli, Staphylococcus species, Streptococcus species, and Pseudomonas species [16][17][18].
Antimicrobial resistance is one of the major problems facing global health today [19], and it is a major source of morbidity and mortality globally [20]. Nowadays it is becoming the leading cause of death globally [21]. A large number of bacteria have acquired and developed antimicrobial resistance mechanisms [20], which constitutes a burden on the global health system with the increasing financial cost [22]. With the increase in drug resistance, there are very few alternatives for patients, and as a result, the number of deaths associated with it has increased [23]. In America, there are 23,000 deaths annually related to drug resistance [20]. The emergence of infectious diseases and the development of antibiotic resistance in bacteria resulting in decreased action or failure of existing antibacterial agents [24], has resulted in an urgent need for the discovery of new, safe, and efficient antibacterial agents [25]. Compounds derived from plants can provide an essential source of new types of antibiotics. There are many types of phytochemicals of plant extract that can exert potential activity on sensitive and multidrug-resistant bacteria [24,26].
Although a number of studies have reported various neem bioactive compounds, there hasn't been much focus on.beta.d-Mannofuranoside, O-geranyl. Scanty information is available on this compound. In a study conducted in India, the authors documented the presence of this compound in mangroves associates crude extract with antimicrobial activity [27], and in another study conducted by Iga et al., they used the isomers of synthetic D-mannofuranoside as antiallergic and antiinflammatory agents [28]. The present study focused on the extraction, GC-MS analysis, and investigation of the antimicrobial activity of crude neem methanolic extract on drug-resistant and sensitive bacteria. Subsequently, the molecular docking and MD simulation studies were explored for the evaluation of the activity of GC-MSidentified compounds on different bacterial essential proteins. Accordingly, in vitro study was explored for the analysis of the antibacterial activity of the pure.beta.d-Mannofuranoside, O-geranyl compound.

Molecular docking
The validation of docking protocols revealed the same orientation of redocked inhibitors with experimentally determined positions, the Root Mean Square Deviation S. epidermidis (n = 1)

Molecular Dynamics simulation
For predicting the interaction stability of ligand and protein complexes, a 50 ns simulation run was performed for the complex of.beta.d-Mannofuranoside, O-geranyl and tyrosyl-tRNA synthetase (PDB ID: 1JIJ), which showed the highest stability during the whole simulation time. The ligand was aligned with protein from the first 5 ns simulation until the end (50 ns) (Fig. 4A). The strongest molecular interaction between ligand and O-geranyl and tyrosyl-tRNA synthetase amino residues was found with Asp117 and Tyr36 (  Phe450, Lys340, and Asn397 of the PBP2X had the most stable interaction during the simulation time (Fig. 5C). The complex of PBP4 (PDB ID: 1TVF) and.beta.d-Mannofuranoside, O-geranyl was stable within 1.2 Å, from 2.4-3.6 Å, this stability occurred after 15 ns simulation time (Fig. 4D). Three residues (Glu83, Ser116, and Ser262) of PBP4 protein showed more than 70% stable interaction with the ligand during the simulation period (Fig. 5D).

Drug likeness, ADME, and Toxicity prediction
In silico tool is used for prediction of.beta.d-Mannofuranoside, O-geranyl drug-likeness, absorption and distribution, Ames test, and carcinogenicity. As shown in Table 4, the compound showed some side effects with remarkable potential to be developed as an antibacterial agent.
The methanolic extract of neem leaves used in this study had shown a potent antibacterial activity on showed interaction with reference co-crystallized ligands (white) and re-docked the same ligands (violet). A. 1JIJ and SB-239629 inhibitor. B. 3TTZ and 07 N inhibitor and C. 5OJ0 and cefepime different bacteria types. The antibacterial activity of crude neem extract was reported previously in many studies worldwide [40][41][42]. Methanol can extract a broad polarity range of compounds with antimicrobial activity [43,43]. In this study, the activity of neem extract was observed on different uropathogens including S. aureus, E. coli, K. pneumoniae, Citrobacter spp., E. faecalis, P. aeruginosa, Proteus spp., and S. epidermidis. The activity of neem leaves ethanolic extract on urine isolates was documented in Sudan [18,45], Pakistan [46], and India [47]. Additionally, Okemo et al. [48] and Pokhrel et al. [40] reported that the crude extract of the neem plant was very effective against S. aureus and E. coli. Our results are different from those obtained by Francine and his colleagues [49] in Rwanda. They reported the activity of neem methanolic extract only on S. aureus but not on E. coli. This variation could be due to differences in neem plants' active constituents and due to differences in environment, genetic factors, and climates [50,51]. The lowest concentration of neem extract reported with bacterial activity was 3.125%, which showed zones of inhibition of more than 10 mm on P. aeruginosa, K. pneumoniae, Citrobacter spp., and E. coli, and 8 mm on Proteus spp. and Staphylococcus spp. These findings are better than Faujdar et al. [47], who also used the same concentrations of neem methanolic extract as in our study, and reported at 6.25 mg/dl concentration a 7 mm zone of inhibition on E. coli and Proteus spp, and 0 mm on P. aeruginosa. Our finding is in concordance with a previous study conducted in Sudan, in which the 6.25 mg/dl concentration was more active on P. aeruginosa, and K. pneumoniae [45]. This could be due to the presence of the same phytoconstituents in neem plants grown in our environment [18,50,51].
Our study showed a high activity of neem extract on bacteria resistant to B-lactam, quinolones, and aminoglycosides, which is consistent with previous findings [45,47]. Although neem was active on pathogenic E. coli and S. aureus, we noticed better activity on the control strains of E. coli and S. aureus than on the pathogenic ones. This could be due to the presence of resistance mechanisms in pathogenic bacteria such as efflux pumps [52,53].
The molecular docking study showed that.beta.d-Mannofuranoside, O-geranyl had potent activity on essential bacterial proteins. The.beta.d-Mannofuranoside, O-geranyl showed hydrogen bonds with residues that are closely interacted with active sites [54]. The Tyr36, Asp40, Tyr170, and Asp177 of S. aureus tyrosyl-tRNA synthetase protein was documented to also have hydrogen bonds with the known co-crystallized protein inhibitor (SB-239629). This activity is concise with our in vitro study and with another in vitro study, in which the authors identified mangrove associates extract with good antibacterial activity, the GC-MS analysis of this extract revealed the presence of a high concentration of.beta.d-Mannofuranoside, O-geranyl [27].
We further evaluated the docking complexes' stability using molecular dynamic simulation, which showed the stability of these complexes during simulation time (50 ns). The ligand was aligned with protein backbones, with a fluctuation of less than 3 Å at most of the simulation period. Usually, changes of the order of 1-3 Å are perfectly acceptable [55]. During the simulation, time residues (Lys340, Trp374, Asn397, and Gln452) of penicillin-binding protein 2X (PDB ID: 5OJ0) showed interaction with protein inhibitor (cefepime) showed a stable interaction with.beta. d-Mannofuranoside, O-geranyl, indicates stability is maintained [56]. On the other hand, the active compounds of neem (.beta.d-Mannofuranoside, O-geranyl) showed a stable interaction with S. aureus penicillin-binding protein 4 (PDB ID: 1TVF); the compound formed a stable water bridge with the catalytic residue of the SXXK motif of the penicillin-binding protein 4 [57].

Conclusion
Methanolic extract of A. indica leaves revealed a potent antimicrobial activity to different types of gram-negative, gram-positive, and control strains. In vitro and in silico experiments revealed the.beta. d-Mannofuranoside, O-geranyl is the most active compound on control strains and different bacterial essential proteins. Using in silico ADME/T prediction, the compound passed 5 rules of drug-likeness properties, so it could be further processed for animal testing and clinical trials for its possible use as an antibacterial agent with commercial values. Moreover, the GC-MS analysis of neem extract revealed the presence of a large number of bioactive components, and the most common were fatty acids (11), hydrocarbons (9), pyridine derivatives (2), and aldehydes (2), which played different biological activities in addition to their nutritional benefits. Based on these findings, the neem plant could help us produce safe and effective medications for a variety of diseases. More in-depth research into these identified phytochemicals will aid pharmaceutical explorations.

Collection and identification of bacterial isolates
A total of 130 urine samples indicated for urine culture and sensitivity testing were collected randomly from patients at East Nile Hospital and Ribat University Hospital in Khartoum state, from January to April 2017. The urine samples were cultured on Cystine Lactose Electrolyte Deficient (CLED) agar media (Hi-Media laboratories PV + Ltd, India), and the clinical isolates were identified using conventional biochemical tests [58].

Plant collection and extraction
Fresh leaves of wild neem were collected from Algazira (Alkamleen city) in central Sudan in March 2017. Leaves were collected from the same tree into clean, dry, labeled plastic bags. Samples were kept frozen at -80 °C until the time of their use [60]. The collected leaves were washed and rinsed to remove dust and other impurities. They were then air-dried and then a total of 50 g of leaves were grounded using a mortar and pestle (Supplementary figure S1), 80% methanol was then used to soak the leaves for three days with daily filtration and evaporation. Then by using a rotary evaporator apparatus under reduced pressure, the solvent was evaporated to dryness [61].

Antibacterial activity of neem extract
The antibacterial activity of neem leaves was tested using the agar well diffusion method on Muller Hinton Agar (MHA) medium against the isolated bacteria and control strains (S. aureus (ATCC 25,923) and E. coli (ATCC 25,922)). Three colonies with similar features were dissolved in 1 mL normal saline and turbidity adjusted to 0.5 McFarland. The isolates were then streaked on the surface of the MHA plate with a sterile swab. Using a cork borer, 6 mm wells were created aseptically on MHA. At sterile conditions, 100 µl of each 50, 25, 12.5, 6.25, 3.125, and 1.5% concentrations of neem extract were poured into media wells [62]. The plates were placed refrigerated for 1 h to allow for extract diffusion before being incubated at 37 °C for 24 h. Methanol alone was used as a negative control. The zone of inhibition was measured (in mm), and the mean was calculated [63]. Three replicates were carried out for the activity of extracted neem against tested organisms. Then the data were presented as mean and standard deviation.

Phytochemical screening of A. indica (neem) extract
The GM-MS method was used to conduct a qualitative and quantitative characterization of neem extract, using the model (GC-MS-QP2010-Ultra) from Shimadzu Company, Japan, with a capillary column Rtx ® -5MS column (30 m, 0.25 mm, 0.25 µm) [64]. The split mode was used for sample injection, and operated in electron ionization (EI) mode at 70 eV, inflow rate of 1.69 ml/min.
Helium gas was used as carrier gas. The injector temperature was set at 300 °C, the temperature of the ion source was 200 °C, and 250 °C was used as interface temperature. The oven temperature program was as follows: the initial temperature at 50 °C rising at 7 °C /min to 180 °C, then the rate changed 10 °C/min reaching the final temperature at 280 °C with 2 min as hold time. In a total 22 min run, the sample was analyzed by the scan mode in a range of 40 to 500 m/z charges to ratio. The neem extract's components were identified by comparing the retention times and mass fragmentation patents with the National Institute of Standards and Technology (NIST) library, and then the results were recorded [65,66].

Molecular docking Proteins selection and preparation
The crystal structures of four essential bacterial proteins were obtained from RCSB PDB database [67] according to their essential role in bacterial cell wall synthesis and protein production in most of our studied isolates, and according to published data [68][69][70][71]. These proteins were Staphylococcus aureus tyrosyl-tRNA synthetase (PDB ID: 1JIJ), DNA gyrase (PDB ID: 3TTZ), Penicillin-Binding Protein 2X (PBP2X) from Streptococcus pneumoniae (PDB ID: 5OJ0), and penicillin-binding protein 4 (PBP4) from Staphylococcus aureus (PDB ID: 1TVF). The proteins' 3D structures were prepared with the Protein Preparation Wizard in Maestro using the default setting. For validation of the docking method, the cocrystalized ligands, SB-239629, 07 N, and cefepime, with their respective structures (1JIJ, 3TTZ, and 5OJ0, respectively), were redocked again using Maestro software [72].

Ligands preparation
The structures of compounds identified by GC-MS (Table 3) were obtained from NCBI PubChem and ChemSpider databases. The ligand's energy was minimized using LigPrep (Schrodinger software, version 2020-3).

Molecular docking
Proteins' active sites were predicted using the Receptor Grid Generation module in Schrodinger. The grids were specified around the co-crystalized ligands or using the SiteMap module to predict S. aureus (PDB ID: 1TVF) protein according to published data [69]. The prepared molecules were docked on protein active sites using extra precision (XP) docking of Schrödinger Maestro software [72]. Ligands were set flexible while proteins were set rigid.

Molecular Dynamic (MD) simulation
Desmond package in the Schrödinger Maestro software [72] was used for MD simulation. The complexes with the best interaction and docking energy were first solvated into the TIP3P water model, an orthorhombic box with boundary 10 Å beyond any of the complex's atoms. Charges were neutralized and OPLS3e force field was used. The particle mesh Ewald method was used for the calculation of long-range electrostatic interactions [73] and cutoff of 12 Å. The molecular dynamic simulation was done in the NPT ensemble at a temperature of 300 K and 1.013 bar pressure over 50 ns and 30 ps for trajectory and 100 ps relaxation time. The trajectories were recorded in 50 ps intervals. After job completion, Root Mean Square Deviation (RMSD) and Root Mean Square Fluctuation (RMSF) were used to examine complexes' stability.

Minimum Inhibitory Concentration (MIC)
Antibacterial activity of the most stable ligand (.beta.d-Mannofuranoside, O-geranyl) from the in silico study was evaluated using in vitro method. The compound was purchased from Apollo Scientific (UK), the compound ID: MolPort-019-937-357, purity 95%, and the Molecular weight was 316.394. The MIC of the compound was evaluated against S. aureus (ATCC 25,923) and E. coli (ATCC 25,922) using the microtitre broth dilution method [74]. A twofold serial dilution of the compound was prepared in broth media (Muller-Hinton), using 96-well microplates flat-bottom plates. One 100 μL of culture media containing bacterial growth adjusted to 5-10 5 CFU/ml was poured into each well. The MIC of the compound was determined at a concentration ranging from 0.5 to 256 μg/ml [75].