The ultraviolet–visible spectral examination was performed using a UV spectrophotometer (Shimadzu Corporation, Japan) to track the transformation of the zinc ion to zinc oxide nanoparticles. The UV spectrum was recorded between 200 and 800 nm.

To characterize the functional groups that participated in ZnONPs formation, Fourier transform infrared (FT-IR) analysis was conducted utilizing a FT-IR 6100 spectrometer (Jasco, Japan) operating at a temperature of 25°C, in the domain from 4,000 to 400 cm⁻¹. FT-IR spectroscopy is a potent analytical technique used in nanoparticle characterization because of its ability to identify individual chemical bonds and functional groups in the sample. FT-IR spectroscopy sheds light on molecular structure, composition, and interactions by evaluating the sample’s absorption or emission of infrared radiation. During the analysis, the FT-IR spectrometer fired infrared light at the ZnONPs sample, and the resulting spectrum was recorded. Peaks and patterns in the spectrum corresponded to typical vibrations of chemical bonds within the nanoparticles, such as stretching and bending modes of functional groups such as -OH (hydroxyl), C=O (carbonyl), C-H (alkyl), and others.

The constancy and charges of the synthesized nanoparticles were assessed using a zeta sizer nano-z’s laser diffractometer (Malvern, United Kingdom). The zeta sizer nano-z’s is equipped with dynamic light scattering (DLS) technology and operates at a temperature of 25°C. Zeta potential measures the electrostatic repulsion forces inside nanostructures and is an important metric for evaluating colloidal stability.

X-ray diffraction (XRD) was carried out using a Bruker D8 Advance Diffractometer (Bruker AXS, Germany) with Cu Ka radiation (k = 1.54) to obtain the XRD pattern of ZnONPs over a 2-theta range of 10–90. XRD detecting and verifying the crystal structure of nanoscale materials, XRD is helpful. The arrangement of atoms in the crystalline lattice can be inferred from the diffraction patterns produced by X-ray diffraction (XRD), which provides important details on the characteristics of the material. The XRD experiment measured a 2-theta range of 10° to 90°. This broad range enabled a complete assessment of the diffraction configurations generated by the ZnONPs, allowing the study of the crystallographic characteristics. X-ray diffraction is operated based on Bragg’s law, with X-rays impacted on a crystalline material diffracted at certain angles given by the spacing of atomic planes inside the crystal lattice. The resulting diffraction pattern contains information on the atoms’ arrangement in the crystal structure, such as lattice parameters, crystal orientation, and phase purity. Using the XRD pattern acquired from the ZnONPs sample, researchers were able to determine the crystal phases present, their relative abundance, and the degree of crystallinity.

Finally, the particle dimension and shape of the zinc oxide nanoparticles were examined utilizing transmission electron microscopy (TEM) with a JEOL-JEM-1011 instrument (Japan). Drops of the suspended nanoparticle solution were located on a carbon-coated copper grid, and the solvent was permitted to outlet gradually before the TEM picture was recorded. Additionally, Furthermore, the morphology of the biosynthesized zinc oxide nanoparticles was further studied using scanning electron microscopy (SEM; Quanta FEG-250, Netherlands).

The antimicrobial activity of the crude extract and ZnONPs was evaluated vs. some pathogenic microbial strains Staphylococcus aureus, Bacillus subtilis, Escherichia coli, Pseudomonas aeruginosa, Candida albicans, and Aspergillus flavus according to the reported procedures (Madkour et al., 2017; Khalaf et al., 2019). The tested samples were dissolved in methanol and a solution of the concentration 1 mg /ml was prepared separately. Paper discs of Whatman filter paper were prepared in standard size (5 cm) and were cut and sterilized in an autoclave. The paper discs soaked in the desired concentration of the complex solution were placed aseptically in the petri dishes containing nutrient agar media (agar 20 g + beef extract 3 g + peptone 5 g) seeded with the tested microbial strains. The petri dishes were incubated at 36°C and the inhibition zones were recorded after 24 h of incubation. Each treatment was replicated three times. Ampicillin and Clotrimazole were utilized as standard antibiotics. The activity of the standard antibiotics was also recorded using the above-mentioned procedures at the same concentration and solvents. The % efficacy index was estimated using the formula:

The cytotoxic activity of the tested samples was performed using MTT assay according to the reported procedures (Saad et al., 2017; Ghareeb et al., 2020) using five cell lines namely; human lung fibroblast (WI38), colorectal carcinoma colon cancer (HCT-116), mammary gland breast cancer (MCF-7) and hepatocellular carcinoma (HEPG-2). Doxorubicin was utilized as a standard anticancer drug. The cell lines were obtained from ATCC via Holding company for biological products and vaccines (VACSERA), Cairo, Egypt. This assay is based on the conversion of the yellow tetrazolium bromide (MTT) to a purple formazan derivative by mitochondrial succinate dehydrogenase in viable cells. Cell lines were cultured in RPMI-1640 medium with 10% fetal bovine serum. Antibiotics added were 100 units/mL penicillin and 100 μg/mL streptomycin at 37°C in a 5% CO₂ incubator. The cell lines were seeded in a 96-well plate at a density of 1.0 × 10⁴ cells/well. at 37°C for 48 h under 5% CO₂. After incubation, the cells were treated with different concentrations of samples and incubated for 24 h. After 24 h of drug treatment, 20 μL of MTT solution at 5 mg/mL was added and incubated for 4 h. Dimethyl sulfoxide (DMSO) in a volume of 100 μL is added into each well to dissolve the purple formazan formed. The colorimetric assay is measured and recorded at the absorbance of 570 nm using a plate reader (EXL 800, United States). The relative cell viability in percentage was calculated as (A570 of treated samples/A570 of untreated sample) × 100.

UPLC analysis was performed via using Shimadzu ExionLC system with the following conditions “Mobile phase A: 0.1% formic acid in water and mobile phase B: acetonitrile; Column: GL-Science (100*2.1) mm, 3 μm; Flow rate: 0.35 mL \min; Column oven: 50°C; Time gradient/Mobile phase B %: 0.0/5.0, 5.0/5.0, 25.0/95.0, 30.0/95.0, 32.0/5.0, and 40.0/5.0. The mass instrument specifications are “Instrument Model: X500 QTOF, Source name: TurboIonSpray, Curtain gas (psi): 30, Ion source gas 1 (psi): 50, Ion source gas 2 (psi): 50, Temperature (°C): 500, Scan type: Full scan -SWATH-Screening, Ion spray voltage (V): Negative, −4,000 V, and CAD gas: 7 (Mahmoud et al., 2023).

The putative target characterization was achieved via Pharmacophore-based Virtual screening using PharmMapper (Wang et al., 2017; El-Sayad et al., 2023). This platform assigns a score to each molecule in the PDB that best fits a pharmacophore model that has been extracted and stored as a library of ligand dataset in mol2 format. After that, when a new molecule is submitted, its fit score is calculated for each pharmacophore, and then each fit score for that pharmacophore is compared to the fit score matrix to determine where it falls on the scale of all the pharmacophore scores. In comparison to chance pharmacophore matching, the pure fit score that results from this procedure carries considerably more weight and assurance. The query structure was submitted to the platform in the PDB format, and the retrieved results were exported as Excel sheet arranging the resulted protein targets according their fit scores.

The crystal structures of E. coli GyrB (PDB ID: 6kzv), and both the human PI3K-γ and c-Src (PDB ID: 2v4l and 3en7, respectively) were used for the docking study using AutoDock Vina (Huey et al., 2012). The co-crystallized ligand in each structure was used to determine the binding site and the docking grid-box in each protein structure, respectively. The co-ordinates of the grid-box were set to be: x = −7.86, y = 16.12, z = 2.49; and x = 45.07, y = 13.12, z = 31.49; and x = −5.09, y = 6.34, z = −6.66, respectively. The ligand to binding site shape matching root means square (RMSD) threshold was set to 2.0 Å. The interaction energies were determined using the Charmm force field (v.1.02) with 10.0 Å as a non-bonded cutoff distance and distance-dependent dielectric. Then, 5.0 Å was set as an energy grid extending from the binding site (Huey et al., 2012). The tested compound retinol was energy minimized inside the selected binding pocket. The editing and visualization of the generated binding poses were performed using Pymol software (Yuan et al., 2017).

NAMD 3.0.0. software was used for performing MDS (Yuan et al., 2017; Ribeiro et al., 2018; Soltane et al., 2023). This software applies the Charmm-36 force field. Protein systems were built using the QwikMD toolkit of the VMD software (Humphrey et al., 1996; Ribeiro et al., 2018), where the protein structures were checked for any missing hydrogens, the protonation states of the amino acid residues were set (pH = 7.4), and the co-crystalized water molecules were removed. Thereafter, the whole structures were embedded in an orthorhombic box of TIP3P water together with 0.15 M Na⁺ and Cl⁻ ions in 20 Å solvent buffer. Afterward, the prepared systems were energy-minimized and equilibrated for 5 ns. The parameters and topologies of the ligands were calculated by using the VMD plugin Force Field Toolkit (ffTK). Afterward, the generated parameters and topology files were loaded to VMD to readily read the protein–ligand complexes without errors and then conduct the simulation steps.

Molecular Mechanics Poisson-Boltzmann Surface Area (MM-PBSA) embedded in the MMPBSA.py module of AMBER18 was utilized to calculate the binding free energy of the docked complex (Humphrey et al., 1996; Ghareeb et al., 2024). 100 frames were processed from the trajectories in total, and the system’s net energy was estimated using the following equation:

Each of the aforementioned terms requires the calculation of multiple energy components, including van der Waals energy, electrostatic energy, internal energy from molecular mechanics, and polar contribution to solvation energy.

The UV–visible spectroscopy was utilized to monitor the green synthesis of ZnONPs using an aqueous fungal extract. Upon addition of the fungal extract to the filtered zinc acetate solution, the color of the solution changed from brownish red to pale yellow, indicating the successful reduction of Zn(CH₃COO)₂ to zinc oxide nanoparticles. The synthesized ZnONPs were then characterized by UV–Vis spectroscopy, which revealed a wide absorption band at 280 and 340 nm as shown in Figure 3, confirming the formation of ZnONPs.

The produced ZnONPs were examined using UV–visible spectroscopy, which is the accepted technique for confirming nanoparticle creation and evaluating their optical characteristics. This method works especially well for “green synthesis,” wherein extracts derived from biological processes serve as reducing agents in metal nanoparticles biosynthesis. The study’s conclusions-which are supported by UV–Vis spectroscopy-highlight the usefulness of the fungal extract in the environmentally friendly synthesis of ZnONPs and call attention to its possible uses in the production of sustainable nanomaterials (Harris et al., 2016).

The FT-IR analysis of the extract and the produced ZnO nanoparticles is shown in Figure 4. Here, FT-IR analysis has been used to identify the functional groups in charge of ZnONPs production and stability. The FT-IR spectra of the green ZnONPs showed that the various functional groups, CH, C=O, C=C, C-O, C-N, and C-C- in the synthesized ZnONPs were connected to peaks at 3500, 3000, 2,800, 1,700, 1,500, 1335.25, 1218.83, and 1030.65 cm⁻¹. The stretching vibration of the hydroxyl group (O-H) is found at 3320, whereas the stretching vibration of the alkane (C-H) is found at peaks at 2919.92 and 2850.58 cm⁻¹. Similarly, a stretch vibration of carbonyl (C=O) was detected at 1,700 cm⁻¹. At 1730.72 cm⁻¹, the aromatic (C=C) stretch bands were observed. Stretching of the C-N was mentioned in the band at 1,200 cm − 1, while the C-O was detected at 1,305. Furthermore, the C-C is located at 1030.65 cm⁻¹. The presence of flavonoids and phenolic acids in the tested extract can be fundamentally linked to the occurrence of absorption bands related to C=C, C=O, C-O, C-N, OH, and CH (Wu et al., 2022). These functional groups contribute to the stability of the prepared ZnONPs as well as the reduction of Zn metal ions.

ZnONPs colloidal solutions displayed constancy in the neutral aqueous system (di-distilled water) since it possesses a zeta potential value of −18.16 mV as presented in Figure 5. ZnONPs colloidal stability in a neutral media has important ramifications for possible uses. Stable colloidal dispersions are crucial for drug delivery in the biomedical domains because they inhibit aggregation and preserve the intended particle size distribution, which influences the kinetics of drug release and the effectiveness of targeting (Wilhelm et al., 2016). Furthermore, in environmental applications, the dispersion and interaction of nanoparticles with contaminants, which facilitates their removal and remediation from aqueous systems, are greatly aided by their colloidal stability. It is important to remember that there are additional variables that might affect colloidal stability than the zeta potential, including particle size, surface charge density, and the presence of stabilizing chemicals (Li et al., 2013).

The X-ray diffraction (XRD) paradigm was utilized to confirm the crystalline nature of ZnONPs. The XRD pattern displayed distinct peaks at (2θ) angles of 31.77, 34.43, 36.26, 47.55, 56.60, 62.88, 66.39, 67.96, and 69.10°, complemented by indicators (100), (002), (101), (102), (210), and (103), sequentially, as noted in Figure 6. The XRD spectra of ZnONPs revealed that the generated nano-ZnO had a spherical form and matched the profile mentioned by Schreyer et al. (2014).

High resolution-TEM was done to investigate the nature of SA-17 ZnONPs particle size distribution and its crystallinity. As shown in Figure 7, the particles consist of a spherical crystal of particle size extending from 3 to 20 nm with an average of 7.2 nm. These spherical crystals contain small particles from 2 to 30 nm with average size of 28.9 nm. The surface morphology of the ZnONPs was explored using a captured SEM image, which showed that the ZnONPs were agglomerated spherical shapes with a diameter ranging from 13 to 55 nm with an average size of 30 nm (Figure 7B). The size detected by SEM was slightly larger than those by TEM as result of surface particles were agglomerated to each other.

The crude extract and biosynthesized ZnONPs showed a pronounced antimicrobial effect toward E. coli, P. aeruginosa, S. aureus, B. subtilis, C. albicans, and A. flavus, respectively. With inhibition percentages ranging from 96.25% to 99.26%, the data showed that Cipro had the highest antibacterial efficiency against all tested pathogens. By contrast, although to a lesser degree, the extract and ZnONPs likewise showed noteworthy antibacterial activity. ZnONPs showed inhibitory percentages ranging from 12.0% to 39.1%, and the extract showed percentages ranging from 28.0% to 52.2%. Significant variations in the antibacterial activity of the tested medicines against every bacterium were found by statistical analysis (p < 0.05). By contrast, the extract and ZnONPs both showed strong antifungal activity. ZnONPs displayed inhibitory percentages ranging from 37.0% to 32.0%, whereas the extract showed percentages ranging from 48.1% to 48.0%. The antifungal activity of the tested medicines against each microbe varied significantly, according to statistical analysis (p < 0.05). The following letters indicate statistical significance for Aspergillus flavus and Candida albicans: Colitrimazole (a) >Extract (b) >ZnONPs (c). The obtained results were compared with standard antibiotics, including Ciprofloxacin and Colitrimazole. Aspergillus species are known for producing unique and diverse secondary metabolites with potent biological activities (Zhang et al., 2018; Abdel-Razek et al., 2020). Figure 8 displayed the inhibition ratio against all tested microbes.

The crude extract demonstrated anticancer activity with IC₅₀ values of (84.55 μM), (31.13 μM), (39.06 mm), and (17.65 μM), against WI38, HCT116, HePG2, and MCF7, respectively. On the other hand, the biosynthesized ZnONPs displayed anticancer activity with IC₅₀ values of (59.74 μM), (43.21 μM), (57.03 mm), and (35.66 μM), against WI38, HCT116, HePG2, and MCF7, respectively. The results were compared with Doxorubicin (Figure 9). The cytotoxicity of the tested drugs against each cell line varied significantly, according to statistical analysis (p < 0.05). The letters that indicate statistical significance for the cell lines WI38, HCT116, HePG2, and MCF7 are as follows: ZnONPs (c) > Extract (b) > Doxorubicin (a). Aspergillus sp. has a great ability to produce a wide of secondary metabolites which are considered a potential source of new anticancer compounds (Kusari et al., 2009; El-Sayed et al., 2021; Noman et al., 2021). A study conducted by Almanaa et al. (2021) reported that A. fumigates extract showed a cytotoxic effect against the HepG-2 cell line with IC₅₀ value of 113 μg/mL. Also, the crude extract of Aspergillus tubenginses ASH4 showed anticancer effect against HCT-116, Hep-G2, and MCF-7 with IC₅₀ values of 9.18, 10.41, and 5.89 μg/mL, respectively (Elkhouly et al., 2021a,b).

A peak demonstrated an [M-H]- m/z at 149 and daughter ions at m/z 105 [M-H-CO₂]-, 87 [M-H-CO₂-H₂O]-, 71 and 59; it was assigned as tartaric acid (McGovern et al., 2009). A peak demonstrated an [M-H]- m/z at 117 and daughter ions at m/z 73 [M-H-CO₂]-, 55 [M-H-CO₂-H₂O]-, and 61; it was assigned as 2-hydroxy-2-methylbutyric acid (John Wiley & Sons, Inc, 2015). A peak demonstrated an [M-H]- m/z at 317 and daughter ions at m/z 273 [M-H-CO₂]-, 165, 113, and 103; it was assigned as 15-Oxo-5Z,8Z,11Z,13E-eicosatetraenoic acid (Sorgi et al., 2018). A peak demonstrated an [M-H]- m/z at 165 and a daughter ion at m/z 119 [M-H-HCOOH]-; it was assigned as arabinonic acid (Moreira et al., 2014). A peak demonstrated an [M-H]- m/z at 115 and a daughter ion at m/z 71 [M-H-CO₂]-; it was assigned as 3-oxopentanoic acid (Kallem et al., 2021). A peak demonstrated an [M-H]- m/z at 137 and daughter ions at m/z 109 [M-H-CO]-, 108 [M-2H-CO]-, 93 [M-H-CO₂]-, 92 [M-2H-CO₂]-, 81, and 65; it was assigned as 3-hydroxybenzoic acid (Ammar et al., 2020; Ghareeb et al., 2023). Hydroxybenzoic acids as a subclass of phenolic acids possess a reliable role in managing neurodegenerative diseases aging, and cancer (e.g., breast cancer; Kalinowska et al., 2021). A peak demonstrated an [M-H]- m/z at 225 and daughter ions at m/z 207 [M-H-H₂O]-, 179, 135, and 97; it was assigned as 2-benzoylbenzoic acid (Zengin et al., 2022). A peak demonstrated an [M-H]- m/z at 167 and daughter ions at m/z 152 [M-H-CH₃]-, 123 [M-H-CO₂]-, 108 [M-H-CH₃-CO₂]-, and 91 [M-H-COOH-OCH₃]-; it was assigned as vanillic acid (Ghareeb et al., 2018a; Sayed et al., 2022). Vanillic acid, an oxidized form of vanillin, is a major active compound isolated from Angelica sinensis and green tea. Vanillin acid is a dietary phenol that can protect biofilms and inhibit lipid peroxidation and eliminates ROS hence act as anti-microbial, anti-inflammatory, anti-cancer factor (Gong et al., 2019). A peak demonstrated an [M-H]- m/z at 183 and daughter ions at m/z 169 [M-H-CH₃]-, 168, 125 [M-H-CO₂]-, 109, and 93; it was assigned as methyl gallate (Ghareeb et al., 2018b, 2019b). A peak demonstrated an [M-H]- m/z at 165 and daughter ions at m/z 121 [M-H-CO₂]-, and 77 [M-H-2CO₂]-; it was assigned as phthalic acid (Gillespie et al., 1989). A peak demonstrated an [M-H]- m/z at 207 and daughter ions at m/z 161 [M-H-C₂H₅OH]-; it was assigned as ethyl caffeate (Barth et al., 2019). A peak demonstrated an [M-H]- m/z at 151 and daughter ions at m/z 136 [M-H-CH₃]-, and 92 [M-H-CH₃-CO₂]-; it was assigned as methyl 3-hydroxybenzoate (Hanley, 2006).

A peak demonstrated an [M-H]- m/z at 191 and daughter ions at m/z 176 [M-H-CH₃]-, 147 [M-H- CO₂]-, 160 [M-H-OH-CH₃]-, 156 [M-H-2OH]-, and 105; it was assigned as 5,7-dihydroxy-4-methylcoumarin (Kerebba et al., 2022). A peak demonstrated an [M-H]- m/z at 207 and daughter ions at m/z 192 [M-H-CH₃]-, 164 [M-H-CH₃-CO]-, and 102; it was assigned as fraxetin (Wang et al., 2016). Fraxetin is a bioactive molecule present in various natural plants, considered as bioactive molecule that acts as anticancer, antioxidative, anti-inflammatory, antidiabetic and antimicrobial activities (Ha and Son, 2024).

A peak demonstrated an [M-H] m/z at 413 and daughter ions at m/z 323 [M-H-90]-, 285 [M-H-128]-, 257 [M-H-156]-, 229 [M-H-184]-; it was assigned as garcinone C (Khaw et al., 2020). Xanthones as secondary metabolites displayed various biological properties, including cytotoxic, antidiabetic, antioxidant, antimicrobial, antitumor, antihypertensive, and anti-inflammatory (Mohamed and Ibrahim, 2022).

A peak demonstrated an [M-H]- m/z at 391 and daughter ions at m/z 203 [M-H-188]-, 187 [M-H-204]-, and 159 [M-H-188-44]-; it was assigned as glabrol (Simons et al., 2009). A peak demonstrated an [M-H]- m/z at 255 and daughter ions at m/z 135 [M-H-C₇H₄O₂]-, and 119 [M-H-C₇H₄O₃]-; it was assigned as 4,2′,5′-trihydroxychalcone (Shan et al., 2017). A peak demonstrated an [M-H]- m/z at 343 and daughter ions at m/z 315 [M-H-CO]-, 313 [M-H-2CH₃]-, 328 [M-H-CH₃]-, 300 [M-H-CH₃-CO]-, and 297 [M-H-H₂O-CO]-; it was assigned as 2′-hydroxy-2,4,5,6′-tetramethoxychalcone (Liu et al., 2016). A peak demonstrated an [M-H]- m/z at 419; it was assigned as pomiferin (Tian et al., 2006). A peak demonstrated an [M-H]- m/z at 625 and daughter ions at m/z 463 [M-H-Glc]⁻, and 301 [M-H-2Glc]-, 271, and 255; it was assigned as quercetin 3,4′-di-O-glucoside (Kumar et al., 2017). A peak demonstrated an [M-H]- m/z at 311 and daughter ions at m/z 293, and 209; it was assigned as 3′,4′-dimethoxy-3-hydroxy-6-methylflavone (Reed, 2009). A peak demonstrated an [M-H]⁻ m/z at 327 and daughter ions at m/z 314, 312, 297, 285, 284, 282, 270, and 112; it was assigned as kaempferol 3,7,4′-trimethyl ether (Dehkordi et al., 2020). A peak demonstrated an [M-H]- m/z at 253 and daughter ions at m/z 238, 223, 179, 151, and 123; it was assigned as 3,6-dihydroxyflavone (March and Brodbelt, 2008). A peak demonstrated an [M-H]- m/z at 253 and daughter ions at m/z 225 [M-H-CO]-, 224 [M-H-CHO]-, 209 [M-H-CO₂]-, 197 [M-H-2CO]-, 185 [M-H-2CO + C]-, and 135 [M-H-C₈H₆O]-; it was assigned as daidzein (Zhao et al., 2018). A peak demonstrated an [M-H]- m/z at 333 and daughter ions at m/z 253 [M-H-SO₃]-, 225 [M-H-SO₃-CO]-, 224 [M-H-SO3-CHO]-, 209 [M-H-SO₃-CO₂]-, 197 [M-H-SO₃-2CO]-, 185 [M-H-SO3-2CO + C]-, and 135 [M-H-SO₃-C₈H₆O]; it was assigned as daidzein 4′-sulfate (Zhao et al., 2018). A peak demonstrated an [M-H]- m/z at 595 and daughter ions at m/z 433 [M-H-Glc]-, 287 [M-H-Glc-Rha]-; it was assigned as eriocitrin (Li et al., 2020).

A peak demonstrated an [M-H]- m/z at 165 and daughter ions at m/z 150 [M-H-CH₃]-, 137 [M-H-CO]-, 122 [M-H-CO-CH₃]-, and 107 [M-H-CO-2CH₃]-; it was assigned as 2′-hydroxy-6′-methoxyacetophenone (Martin, 2013).

A peak demonstrated an [M-H]- m/z at 239 and daughter ions at m/z 221 [M-H-H₂O]-, 193 [M-H-CH₂O₂]-, 179 [M-H-C₂H₄O₂]-, 165 [M-H-C₃H₆O₂]-, and 138 [M-H-C₄H₇NO₂]-; it was assigned as β-carboline-1-propionic acid (Chua et al., 2011). A peak demonstrated an [M-H]- m/z at 144 and daughter ions at m/z 114, and; it was assigned as quinolin-8-ol (Clugston, 1996).

A peak demonstrated an [M-H]- m/z at 121 and daughter ions at m/z 93 [M-H-CO]- and 92 [M-H-CHO]-; it was assigned as 3-hydroxybenzaldehyde (Castillo et al., 2023).

A peak demonstrated an [M-H]- m/z at 241 and daughter ions at m/z 225, 197, 181, and 169; it was assigned as pinostilbene (Chen et al., 2016).

A peak demonstrated an [M-H]- m/z at 493 and daughter ions at m/z 363 [M-H-cinnamoyl]-, 345 [M-H-cinnamoyl-H₂O]-, 201 [M-H-cinnamoyl-Glc]-, 183 [M-H-cinnamoyl-H₂O-Glc]-, and 147 [M-H-C₁₅H₂₂O₉]-; it was assigned as harpagoside (Xiong et al., 2010).

A peak demonstrated an [M-H]- m/z at 269 and daughter ions at m/z 251 [M-H-H₂O]-, 241 [M-H-CO]-, 227 [M-H-CO-CH₂]-, 225, 223 [M-H-CO-H₂O]-, 195 [M-H-CO-H₂O-CO]-; it was assigned as emodin (Fu et al., 2015). A peak demonstrated an [M-H]- m/z at 283 and daughter ions at m/z 268 [M-H-CH₃]-, 240 [M-H-CH₃-CO]-, 212, and 184; it was assigned as physcion (Fu et al., 2015).

A peak demonstrated an [M-H]- m/z at 265 and a daughter ion at m/z 224 [M-CH₂CH=CH₂-H]-; it was assigned as honokiol (Wu et al., 2006).

To find out how crude extract exerts its antibacterial and anticancer activities, all the modeled structures of the UPLC–MS-annotated compounds in these samples were subjected to pharmacophore-based virtual using the PharmMapper platform (Wang et al., 2017).

PharmMapper can screen and recommend the most probable protein targets of a query molecule based on its pharmacophore model by mapping its fundamental pharmacophore characteristics (i.e., the spatial arrangement of structural features).

Accordingly, the compounds that are compatible with these pharmacophore maps have a greater potential for binding to the same protein targets. Therefore, the annotated compounds (Table 2) were run through PharmMapper to find their potential protein targets. The obtained results were arranged according to their degree of conformity to the pre-determined parameters (the Fit score). Only bacterial-relevant and cancer-relevant targets were chosen.

As a result, DNA gyrase subunit-B (GyrB) of E. coli (PDB ID: 6KZV) was found to be the top-scoring bacterial-relevant hit for glabrol and pomiferin (Figure 11). Hence, these metabolites can be considered tentatively as the key antibacterial metabolites in the tested extract.

On the other hand, Daidzein 4′-sulfate (Figure 11) was predicted to bind with the active site of human phosphoinositide 3-kinase gamma (PI3K-γ; PDB ID: 2V4L) with a Fit score of 8.28.

It is well-known that GyrB is an essential protein that provides the necessary energy for the GyrA subunit, which in turn unfolds the bacterial DNA, making it available for the DNA replicating enzyme, and thus blocking these enzymes (i.e., GyrA or GyrB) will stop the bacterial DNA replication and eventually bacterial death (Wigley et al., 1991; Ushiyama et al., 2020).

Regarding the cancer-relevant protein, PI3K-γ has been reported in many previous reports to be over-expressed in different types of human tumors, particularly breast cancers (Apsel et al., 2008; Kwak et al., 2019; Miricescu et al., 2020).

To investigate the binding modes of each aforementioned compound (Figure 11) with GyrB and PI3K, their modeled structures were prepared and re-docked into the active sites of each protein. Thereafter, the resulting binding poses were subjected to 50 ns-long MD simulation runs to test the binding affinity and stability of each structure inside the active sites of the suggested protein targets. First, the re-docking of glabrol and pomiferin structures into the GyrB’s active site achieved binding modes and docking scores comparable to those of the co-crystalized inhibitor (Figure 11; Table 2).

As shown in Figures 10A–C, the structures of glabrol and pomiferin were able to achieve binding modes convergent to that of the co-crystalized inhibitor forming comparable hydrophilic and hydrophobic interactions (Table 2). H-bonds with ASP-73 and ARG-76 were the common hydrophilic interactions among the three structures alongside the co-crystalized inhibitor (Figure 12). Second, re-docking of daidzein 4′-sulfate structure into the active sites of PI3K resulted in binding mode and docking score comparable to those of the co-crystalized inhibitor (Figure 13; Table 3). Subsequent experiments using molecular dynamic (MD) simulation (lasting 50 ns) showed that glabrol and pomiferin structures were able to achieve stable binding modes inside the GyrB’s active site, with relative mean square deviations (RMSDs) of 2.523 Å and 1.765 Å, respectively (Figure 14). Hence, the calculated binding free energies (ΔG_Binding) these structures in comparison with that of the co-crystallized inhibitor were convergent (ΔG_Binding = −7.57, −8.54, and −8.76 kcal/mol, respectively).

Regarding the binding behavior of daidzein 4′-sulfate structure inside the active sites of PI3K throughout 50 ns-long MD simulations in comparison with the co-crystallized inhibitor, it also achieved stable binding concerning the co-crystallized inhibitor with an average RMSD of 2.552 Å. Additionally, their calculated ΔG_Binding values were convergent to that of the co-crystalized inhibitor (ΔG_Binding = −8.96 and −8.87 kcal/mol for daidzein 4′-sulfate and the co-crystallized inhibitor, respectively; Figure 15).

Moreover, modeling and MD simulation results suggest that glabrol and pomiferin in the tested extract are the antibacterial-relevant metabolites, while daidzein 4′-sulfate is the anticancer-relevant metabolite. In previous reports, glabrol and pomiferin have been shown to exert interesting antibacterial efficacies vs. both Gram-positive and Gram-negative pathogenic bacteria including antibiotics-resistant ones (Gerhäuser, 2005; Stompor and Żarowska, 2016; Wu et al., 2019; Mohamed et al., 2022).