Introduction

Despite significant advancements in science and technology, several illnesses continue to afflict us and threaten our lives. Among these, the rapid emergence of antimicrobial resistance (AMR) in both Gram-positive and Gram-negative pathogens has significantly reduced the effectiveness of conventional antibiotics, creating an urgent need for alternative therapeutic strategies. In the pursuit of comprehending and addressing ailments, humans have identified several plants with medicinal properties1. Many of these medicinal plants have been utilized for millennia by a substantial segment of the population and continue to be employed in healthcare, either independently or in conjunction with contemporary pharmaceuticals2. Conventional medicinal herbs are favored by individuals in underdeveloped nations owing to their perceived safety, accessibility, and affordability. Importantly, numerous medicinal plants contain bioactive secondary metabolites—such as phenolics, flavonoids, terpenoids, and alkaloids—that have demonstrated antimicrobial activity (Verma, Sinha et al., 2021,3.

Interest in microbicidal plants is steadily increasing because herbal pesticides and fungicides are considered harmless4 and easily biodegradable5. Origanum majorana has been reported to possess antibacterial activity6,7. The in vitro activity of the methanol extract was verified against six bacteria and seven fungi, and was found to be an effective herbal protective against various pathogenic microorganisms. High toxicity against Aspergillus niger has also been reported4. The antimicrobial activity of O. majorana is largely attributed to its essential oils and phenolic constituents, which may disrupt microbial cell membranes and enzymatic systems8.

Costus speciosus possesses diverse pharmacological activities (Zishan, Uddin et al., 2024). Traditionally, its rhizomes are used to cure pneumonia and skin diseases, conditions frequently associated with microbial infections, whereas its leaves are employed in the management of mental disorders (Zishan, Uddin et al., 2024). Recent studies have identified the rhizome extract as the most active part of the plant, with antimicrobial activity comparable to that of the standard antibiotic, gentamycin. A simple phytochemical analysis was conducted to estimate total phenols, ortho-dihydric phenols, and the alkaloid diosgenin in the rhizomes at both the vegetative and fruiting stages using spectroscopy, with the aim of assessing disease resistance properties9. The use of C. speciosus rhizome extract as a potential bactericide for the cure and prevention of bacterial infections has been suggested10. The presence of steroidal saponins such as diosgenin further supports its potential antimicrobial relevance11.

Chatoui12, reported that methanol and ethyl acetate extracts of Lepidium sativum seeds displayed strong antimicrobial activity against all tested bacteria and exhibited noteworthy action against Rhodococcus equi (Chatoui, Talbaoui et al., 2016). Furthermore, the antibacterial effects of aqueous and ethanolic extracts as well as juice from L. sativum were evaluated against both Gram-negative and Gram-positive bacteria (Staphylococcus aureus, Klebsiella pneumoniae, Proteus, Pseudomonas aeruginosa, and Streptococcus mutans)13. The plant extracts inhibited all bacteria under study, except K. pneumoniae, whereas the juice showed no antibacterial effect. The minimum inhibitory concentration(MIC) of L. sativum extract was 3% for Proteus and K. pneumoniae, whereas other bacterial species were susceptible to all tested dilutions13,14. These findings indicate solvent-dependent variability in antimicrobial efficacy, suggesting that extraction methodology critically influences biological activity15.

Although individual studies have evaluated these plants separately, limited research has comparatively examined their antimicrobial potential under standardized extraction conditions or integrated experimental findings with computational mechanistic analysis. Furthermore, most previous investigations have relied primarily on conventional solvents such as ethanol, methanol, or distilled water, with minimal exploration of alternative extraction media16.

Most researchers investigating the antimicrobial properties of medicinal plants commonly use ethanol, methanol, or distilled water as extraction solvents, whereas few have explored the potential of Zamzam water in this context17. Zamzam water possesses a distinctive mineral composition, including elevated levels of calcium, magnesium, and bicarbonates, which may influence extraction efficiency by altering solvent polarity, ionic strength, and phytochemical solubility. Such physicochemical properties could enhance the recovery of certain polar bioactive compounds compared with distilled water18.

The four plants selected in this study—Lepidium sativum, Costus speciosus, Origanum majorana, and Linum usitatissimum—represent diverse phytochemical classes, including essential oils, flavonoids, lignans, and steroidal saponins. Evaluating them collectively enables comparative assessment across chemically distinct plant matrices and may provide broader insight into solvent-dependent antimicrobial activity19.

In light of increasing antibiotic resistance, this study aimed to identify plant-based alternatives for infection treatment. The primary objective of this research was to assess the antibacterial and antifungal efficacy of these four plants against clinically relevant microbes using the agar diffusion and microbroth dilution methods. We hypothesized that solvent type, including mineral-rich Zamzam water, influences phytochemical extraction efficiency and antimicrobial activity, and that selected phytocompounds would demonstrate measurable binding affinity toward essential bacterial enzymes20.

Materials and methods

Materials and plants

The apparatus used included different instruments, glassware, chemicals, and reagents. All items, reagents, glassware, and chemicals used were pure and standard. Plant specimens were collected from various locations in southern Saudi Arabia (Jazan). The plants under test were collected in June 2024 and stored at room temperature, whereas the fixed oils were stored in a refrigerator. The four plant specimens and two fixed oils are listed in Table 1. The plant specimens were collected from Al-Darb region, Jazan Province, Saudi Arabia (16.7396° N, 42.2569° E). All plant materials were carefully examined for identification21,22,23. The Costus speciosus used in our study was formally identified by Dr. Mukul Sharma, Department of Botany, Jazan University, and a voucher specimen has been deposited at the Health Research Centre Herbarium of Jazan University under the accession number [JUH2025-CS01]. The collection of plant specimens was conducted with the necessary permissions obtained from the Jazan University24.

Table 1 Plant materials.
Full size table

Preparation of plant extracts using various solvents

Different plant extracts were prepared using ethanol, petroleum ether, methanol, distilled water, and Zamzam water. The plant materials were air-dried and ground into a fine powder. One hundred grams of each powdered plant material were soaked in one liter of the respective solvent in 1000 mL conical flasks, which were then stoppered and left to stand for 72 h to allow complete extraction of active constituents25. The extracts were filtered through filter paper, and the solvents were removed using a rotary evaporator under reduced pressure to obtain concentrated extracts. The dried extracts were weighed, and the extraction yield (%) was calculated as the weight of the dried extract divided by the initial weight of plant material (100 g), multiplied by 100. For Origanum majorana, the extraction yields were 4.7% (ethanol), 4.9% (methanol), 3.2% (petroleum ether), 3.8% (distilled water), and 4.1% (Zamzam water). For Costus speciosus, the yields were 4.5% (ethanol), 4.8% (methanol), 3.4% (petroleum ether), 3.7% (distilled water), and 4.2% (Zamzam water). For Lepidium sativum, the yields were 4.3% (ethanol), 4.6% (methanol), 3.1% (petroleum ether), 3.6% (distilled water), and 4.0% (Zamzam water). For Linum usitatissimum, the yields were 4.6% (ethanol), 4.8% (methanol), 3.3% (petroleum ether), 3.5% (distilled water), and 4.4% (Zamzam water). For aqueous and Zamzam water extractions, the plant powders were macerated in one liter of solvent for 24 h, filtered, evaporated to dryness, and stored in sterile screw-capped vials at 4 °C. Ethanol and petroleum ether extracts were prepared by immersing 100 g of powder in one liter of solvent with intermittent agitation for 72 h, followed by filtration using a Buchner funnel, evaporation, and storage in pre-weighed flasks. The methanol extract was prepared similarly, followed by solvent removal under reduced pressure and vacuum drying of the semisolid mass to obtain the final residue. All dried extracts were stored at 4 °C until further analysis.

Antimicrobial activity

Microorganisms

All microorganisms used in this study were standard organisms from the American Type Culture Collection (ATCC): Staphylococcus aureus (25923), Bacillus subtilis (11774), Proteus vulgaris (33420), Klebsiella pneumoniae (10031), and Candida albicans (10231). Different types of culture media, including nutrient agar, nutrient broth, MacConkey agar, blood agar, Mueller-Hinton agar, Mueller-Hinton broth, plate count agar, and Sabouraud dextrose agar, have been used for the cultivation, testing, and storage of microorganisms26.

Preparation of the standard bacterial suspension

Nutrient agar slopes were aseptically covered with one milliliter aliquots of a 24-hour broth culture of the tested organisms, which were then incubated for 24 h at 37 °C. To create a suspension with approximately 10–10 colony-forming units per milliliter, the bacterial growth was collected, rinsed with sterile normal saline, and then suspended in a tiny volume of normal saline. The suspension was stored in a refrigerator at 4 °C until use. The surface viable counting approach was used to calculate the average number of viable organisms per mL of the stock solution. The stock solution was serially diluted in sterile normal saline, and the corresponding dilutions were deposited onto the surface of dry nutritional agar plates in 0.02 ml quantities (one drop) using a digital pipette (Finn pipette, adjustable volume). After drying for two hours at room temperature, the Petri dishes were incubated for twenty-four hours at 37 °C. The number of colonies formed by each drop was counted following incubation. The dilution factor, which is the number of colony-forming units (CFU/ml) per mL of solution, was calculated by multiplying the average number of colonies per drop (0.02 ml) by 50. All the aforementioned experimental parameters were kept constant whenever a new stock suspension was made to produce suspensions with extremely near viable counts (Jain, Begum et al. 2024).

Antimicrobial testing

Antimicrobial activity was tested using the agar diffusion method (Nguyen, Miyamoto et al., 2024). Two hundred and fifty milliliters of sterilized Mueller-Hinton agar were employed for the testing. The inoculum size for each test bacterium was adjusted to a suspension of 10⁶ cells. Two milliliters of a 24-hour-old culture was added to 250 ml of melted, cooled test agar. After thorough mixing, nearly 20 ml of this seeded medium was transferred into pre-sterilized Petri dishes with a diameter of 10 cm and allowed to congeal. A sterile cork borer was used to bore three wells (10 mm in diameter) in the agar and then remove the agar discs. Each well was filled with a 0.1 ml aliquot of the diluted extract using a pipette. The plate was maintained at ambient temperature for 2 h to allow the extract to diffuse into the agar. The plates were incubated at 37 °C for 24 h. Upon completion of incubation, the zones of inhibition were measured to the nearest millimeter in diameter21. The Activity Index (AI) was calculated according to the formula AI = inhibition zone of the sample/inhibition zone of the standard. Cephalosporin, Gentamycin, Tetracycline, and Clindamycin were used at concentrations ranging from 20 to 10 µg/ml as positive controls, and Dimethyl-Sulphoxide (DMSO) was used as a negative control.

Minimum inhibition concentration (MIC) and minimum bactericidal concentration (MBC)

The MIC values were determined using the broth microdilution method according to Clinical and Laboratory Standards Institute (CLSI) guidelines, with minor modifications. Serial two-fold dilutions of each plant extract (0.156–100 mg/mL) were prepared in Mueller-Hinton broth. The bacterial inoculum was adjusted to 0.5 McFarland standard and diluted to obtain a final concentration of 5 × 10⁵ CFU/mL in each well. Each well contained Mueller-Hinton broth, the appropriate extract dilution, and the standardized bacterial suspension. Growth and sterility controls were included. Tetracycline hydrochloride (0.1 mg/mL) and amoxicillin (0.1 mg/mL) served as positive controls for Staphylococcus aureus and Escherichia coli, respectively. Plates were incubated at 37 °C for 24 h.

The MIC was defined as the lowest concentration showing no visible growth compared with the control and was confirmed by optical density measurement at 600 nm. All assays were performed in triplicate. For MBC determination, aliquots from wells without visible growth were subcultured onto Mueller-Hinton agar plates and incubated for 24 h at 37 °C. The MBC was defined as the lowest concentration producing no colony growth (≥ 99.9% bacterial reduction).

Preparation of antimicrobial stock solutions

Stock antibiotic solutions were prepared based on manufacturer potency specifications. The required weight (W, mg) of each antibiotic was calculated using the equation:

where P represents the potency provided by the manufacturer (µg/mg), C is the desired final concentration (mg/L), and V is the required volume (mL). All stock solutions were freshly prepared and filter-sterilized before use. The concentrations (50 and 100 mg/mL) were selected as commonly used screening doses for crude plant extracts in agar diffusion assays to ensure detectable activity and comparability, while maintaining extract solubility and diffusion. The final concentration of DMSO in all assay wells did not exceed 1% (v/v), and control experiments confirmed that this concentration had no inhibitory effect on microbial growth.

Molecular docking experiment

Software packages

The graphical user interface program FLARE version7.2.0 software Package, CRESSET U.K. (https://cresset-group.com/about/news/flare-v7-released/) was used as the computational tool in this study.

Protein preparation

During the virtual screening process, the selection of target proteins is essential for identifying potential antibacterial compounds. Two main targets representing Gram-negative and Gram-positive bacteria were selected to investigate the principal active molecules that may be involved in the observed antimicrobial activity of the tested extracts at the molecular level. For Gram-negative bacteria, DNA gyrase was chosen as the key target because of its essential role in bacterial DNA replication. The Protein Data Bank (PDB) structure used for DNA gyrase has the PDB ID 1KZN, with a resolution of 2.30 Å. Dihydrofolate reductase (DHFR) was selected as the vital target for Gram-positive bacteria because of its role in bacterial folate metabolism. The PDB ID used for DHFR was 3SRW, with a resolution of 1.70 Å. By targeting these well-established enzymes, the screening process aimed to identify compounds with strong binding affinities that may contribute to novel antibacterial development.

The targets were downloaded and loaded into the Flare software in the PDB format27. Target preparations were conducted in Flare using default settings. All proteins were prepared by adding hydrogen and ensuring that all-atom valences were satisfied, to enable accurate docking simulations. After preparation, the 3D structures of the targets were minimized using the Cresset Flare software with normal-type calculation methodology28.

Ligands preparation

Selected compounds from the literature were collected as follows: 144 compounds related to the fruits of O. majorana (Bouyahya, Chamkhi et al., 2021), 16 compounds from the leaves of L. sativum (Painuli, Quispe et al., 2022), 14 compounds from the seeds of L. usitatissimum29, and 18 compounds from the roots of C. speciosus (Sohrab, Mishra et al., 2021) were downloaded from PubChem as 3D conformer SDF files. The SDF files of the selected compounds from each corresponding plant docked against the specified targets were loaded into Flare software (Cresset) and treated using the default settings. The ligands were prepared by assigning proper bond orders and generating the correct tautomer and/or ionization state, and then minimizing and optimizing within the Flare software using the accurate type calculation method28.

Docking process

The docking calculations were performed in the Flare software using the normal mode and default settings. The active site of the targeted proteins was selected by clicking the ‘picking active site’ button in the Flare software28. The grid box was well-defined according to the co-crystallized ligands (chlorobiocin (A CBN 1) for 1KZN and 7-aryl-2,4-diaminoquinazolines (X Q27 168) for 3SRW). Docking parameters were fixed for all docking runs using default settings. Docking analyses of the selected compounds against each of the assigned target proteins were performed using the Lead Finder algorithm. Lead Finder’s scoring function uses a semi-empirical molecular mechanical function that explicitly accounts for various types of molecular interactions30. After docking, the best binding pose for each phytocompound was selected. These poses were predicted to be the most stable conformations for binding to the active site of the target protein. The best poses are then generated and visualized. Docking protocol validation was performed by re-docking the co-crystallized ligands (chlorobiocin for 1KZN and 7-aryl-2,4-diaminoquinazolines for 3SRW) into their respective binding sites using the same docking parameters. The root mean square deviation (RMSD) between the crystallographic pose and the re-docked pose was calculated. The obtained RMSD values were 1.42 Å for 1KZN and 1.36 Å for 3SRW, both below the acceptable threshold of 2.0 Å, confirming the reliability and accuracy of the docking protocol.

Statistical analysis

Statistical analysis was performed using one-way ANOVA followed by Tukey’s post hoc multiple comparison test to evaluate differences among treatments. Data are presented as mean ± SD (n = 3). Statistical significance was defined at p < 0.05. Analyses were conducted using SPSS software (version 26.0).

Results

The antimicrobial activity of the four plant species extracted with ethanol, methanol, petroleum ether, aqueous, and Zamzam water was evaluated against four bacterial strains (Staphylococcus aureus, Bacillus subtilis, Klebsiella pneumoniae, and Proteus vulgaris) and one fungal strain (Candida albicans). Results are presented as mean ± SD (n = 3), and statistical differences were determined using one-way ANOVA followed by Tukey’s post hoc test (p < 0.05). Inhibition zones were categorized as sensitive (≥ 18 mm), intermediate (14–17 mm), or resistant (< 14 mm) (Table 2).

For S. aureus, the largest inhibition zones were observed with C. speciosus ethanol and methanol extracts at 100 mg/mL (35 ± 1.4 mm), which were statistically comparable to cephalosporin (33 ± 1.3 mm) (p < 0.05). O. majorana methanol extract (100 mg/mL) also demonstrated high activity (33 ± 1.3 mm). Petroleum ether extracts showed reduced or no activity in certain species, particularly L. usitatissimum (0 mm). Against B. subtilis, C. speciosus ethanol and methanol extracts (28 ± 1.1 mm and 27 ± 1.1 mm, respectively) were among the highest-performing treatments and were not statistically different from cephalosporin (30 ± 1.2 mm). Aqueous extracts generally produced moderate inhibition, while petroleum ether extracts of L. usitatissimum showed no activity.

For K. pneumoniae, the highest inhibition zone was recorded for L. usitatissimum ethanol extract (100 mg/mL) (37 ± 1.5 mm), significantly greater than reference antibiotics (p < 0.05). C. speciosus aqueous extract (33 ± 1.3 mm) and O. majorana petroleum ether extract (31 ± 1.2 mm) also demonstrated strong activity. Several petroleum ether extracts showed no inhibition. In the case of P. vulgaris, L. usitatissimum ethanol extract (30 ± 1.2 mm) and O. majorana petroleum ether extract (30 ± 1.2 mm) produced the largest inhibition zones. Tetracycline and clindamycin showed no measurable inhibition against this organism under the tested conditions. For C. albicans, the highest inhibition was observed with O. majorana petroleum ether extract (27 ± 1.1 mm) and L. usitatissimum ethanol extract (26 ± 1.0 mm). Petroleum ether extracts of C. speciosus and L. usitatissimum exhibited no antifungal activity. Across organisms, ethanol and methanol extracts consistently produced inhibition zones within the sensitive range. Zamzam water extracts yielded inhibition zones comparable to alcohol-based extracts in several plant species, particularly L. usitatissimum and C. speciosus.

Table 2 Inhibition Zones (Mean ± SD, mm) with Statistical Grouping. Values sharing different superscript letters within a column are significantly different (p < 0.05).
Full size table

Docking scoring results

Docking against DNA gyrase (PDB ID: 1KZN)

Docking scores of selected phytocompounds against DNA gyrase are presented in Table 3. The co-crystallized ligand clorobiocin exhibited a binding affinity of − 10.112 kcal/mol. β-Caryophyllene from O. majorana (Fig. 1) demonstrated a docking score of − 11.549 kcal/mol, while quercetin and kaempferol derivatives from L. sativum (Fig. 2) showed docking scores ranging from − 10.525 to − 10.039 kcal/mol. Vitexin from L. usitatissimum (Fig. 3) and sitosterol from C. speciosus (Fig. 4) exhibited lower binding affinities (− 8.972 and − 7.149 kcal/mol, respectively). All compounds were predicted to occupy the active binding pocket of DNA gyrase.

Table 3 Docking results against 1KZN against O. majorana, L. usitatissimum, L. sativum, C. speciosus.
Full size table
Fig. 1
Fig. 1The alternative text for this image may have been generated using AI.
Full size image

Molecular visualization of β-Caryophyllene_1KZN interaction. (A) showed zoom out of β-Caryophyllene (blue) interacting and clustered to the same binding site of the co crystalized ligand clorobiocin (green) within the active site of DNA-gyrase with PDB ID 1KZN. (B) showed zoom in of β-Caryophyllene interacting to the same binding site of clorobiocin. (C) 2D map of β-Caryophyllene interacting to crucial amino acids of the active site. (D) Grid visualization showing their overall electrostatic complementary to the active site.

Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

Molecular visualization of Quercetin _1KZN interaction. (A) showed zoom out of quercetin (magenta) interacting and clustered to the same binding site of the co crystalized ligand clorobiocin (green) within the active site of DNA-gyrase with PDB ID 1KZN. (B) showed zoom in of quercetin interacting to the same binding site of clorobiocin. (C) 2D map of quercetin interacting to crucial amino acids of the active site. (D) Grid visualization showing their overall electrostatic complementary to the active site.

Fig. 3
Fig. 3The alternative text for this image may have been generated using AI.
Full size image

Molecular visualization of Vitexin _1KZN interaction. (A) showed zoom out of vitexin (yellow) interacting and clustered to the same binding site of the co crystalized ligand clorobiocin (green) within the active site of DNA-gyrase with PDB ID 1KZN. (B) showed zoom in of vitexin interacting to the same binding site of clorobiocin. (C) 2D map of vitexin interacting to crucial amino acids of the active site. (D) Grid visualization showing their overall electrostatic complementary to the active site.

Fig. 4
Fig. 4The alternative text for this image may have been generated using AI.
Full size image

Molecular visualization of sitosterol _1KZN interaction. (A) showed zoom out of sitosterol (cyan) interacting and clustered to the same binding site of the co crystalized ligand clorobiocin (green) within the active site of DNA-gyrase with PDB ID 1KZN. (B) showed zoom in of sitosterol interacting to the same binding site of clorobiocin. (C) 2D map of sitosterol interacting to crucial amino acids of the active site. (D) Grid visualization showing their overall electrostatic complementary to the active site.

Docking against dihydrofolate reductase (PDB ID: 3SRW)

Docking results against DHFR are shown in Table 4. The co-crystallized ligand (7-aryl-2,4-diaminoquinazoline derivative) displayed a binding affinity of − 6.098 kcal/mol. β-Caryophyllene from O. majorana showed the highest binding affinity (− 13.169 kcal/mol) (Fig. 5). Flavonoids including quercetin, kaempferol, and luteolin derivatives demonstrated docking scores between − 10.755 and − 9.593 kcal/mol (Fig. 6). Compounds from L. usitatissimum ranged between − 10.004 and − 9.402 kcal/mol (Fig. 7), while sitosterol from C. speciosus exhibited − 7.404 kcal/mol (Fig. 8). No causal inference was made between docking scores and antimicrobial potency; docking findings are reported as predictive binding estimations.

Table 4 Docking results of the principal compounds from O. majorana, L. usitatissimum, L. sativum, C. speciosus against 3SRW.
Full size table
Fig. 5
Fig. 5The alternative text for this image may have been generated using AI.
Full size image

Molecular visualization of β-Caryophyllene _3SRW interaction. (A) showed zoom out of β-Caryophyllene (blue) interacting and clustered to the same binding site of the co crystalized ligand 7-aryl-2,4-diaminoquinazolines (green) within the active site of DHFR with PDB ID 3SRW. (B) showed zoom in of β-Caryophyllene interacting to the same binding site of 7-aryl-2,4-diaminoquinazolines. (C) 2D map of β-Caryophyllene interacting to the crucial amino acids of the active site. (D) Grid visualization showing their overall electrostatic complementary to the active site.

Fig. 6
Fig. 6The alternative text for this image may have been generated using AI.
Full size image

Molecular visualization of quercetin _3SRW interaction. (A) showed zoom out of quercetin (magenta) interacting and clustered to the same binding site of the co crystalized ligand 7-aryl-2,4-diaminoquinazolines (green) within the active site of DHFR with PDB ID 3SRW. (B) showed zoom in of quercetin interacting to the same binding site of 7-aryl-2,4-diaminoquinazolines. (C) 2D map of quercetin interacting to the crucial amino acids of the active site. (D) Grid visualization showing their overall electrostatic complementary to the active site.

Fig. 7
Fig. 7The alternative text for this image may have been generated using AI.
Full size image

Molecular visualization of vitexin _3SRW interaction. (A) showed zoom out of vitexin (yellow) interacting and clustered to the same binding site of the co crystalized ligand 7-aryl-2,4-diaminoquinazolines (green) within the active site of DHFR with PDB ID 3SRW. (B) showed zoom in of vitexin interacting to the same binding site of 7-aryl-2,4-diaminoquinazolines. (C) 2D map of vitexin interacting to the crucial amino acids of the active site. (D) Grid visualization showing their overall electrostatic complementary to the active site.

Fig. 8
Fig. 8The alternative text for this image may have been generated using AI.
Full size image

Molecular visualization of sitosterol _3SRW interaction. (A) showed zoom out of sitosterol (cyan) interacting and clustered to the same binding site of the co crystalized ligand 7-aryl-2,4-diaminoquinazolines (green) within the active site of DHFR with PDB ID 3SRW. (B) showed zoom in of sitosterol interacting to the same binding site of 7-aryl-2,4-diaminoquinazolines. (C) 2D map of sitosterol interacting to the crucial amino acids of the active site. (D) Grid visualization showing their overall electrostatic complementary to the active site.

Discussion

This study evaluated the antimicrobial activity of multiple solvent extracts of Origanum majorana, Costus speciosus, Lepidium sativum, and Linum usitatissimum, and explored potential molecular interactions of selected phytocompounds with bacterial targets. The findings indicate that extraction solvent significantly influences antimicrobial performance, with ethanol and methanol extracts consistently demonstrating higher inhibition zones across most tested microorganisms. Similar solvent-dependent antimicrobial activity has been reported previously4,31,32. Zamzam water extracts showed activity comparable to alcohol-based extracts in several cases, exceeding that of distilled water extracts.

The enhanced activity observed with alcohol-based solvents is consistent with their ability to solubilize phenolic compounds, flavonoids, and terpenoids, which are frequently associated with antimicrobial properties (Kwansa-Bentum, Okine et al., 2023). Previous investigations have demonstrated that methanol extracts of O. majorana exhibit strong antifungal and antibacterial effects, particularly against Aspergillus niger and Gram-positive bacteria4. Likewise, C. speciosus rhizome extracts have shown notable antibacterial activity comparable to standard antibiotics5,9. These reports align with the present findings, where alcohol-based extracts demonstrated consistent inhibition against S. aureus and B. subtilis.

Differences between Gram-positive and Gram-negative bacteria were evident. Gram-positive strains (S. aureus and B. subtilis) generally exhibited higher susceptibility to several extracts, which may reflect their thick peptidoglycan cell wall lacking an outer membrane. In contrast, Gram-negative bacteria such as K. pneumoniae and P. vulgaris displayed variable responses, potentially due to the presence of an outer membrane that can limit permeability of antimicrobial agents33. However, certain ethanol extracts, particularly from L. usitatissimum, demonstrated substantial activity against K. pneumoniae, indicating that some phytochemicals may overcome outer membrane barriers.

The observation that Zamzam water extracts frequently showed greater activity than distilled water extracts suggests that solvent composition may influence extraction efficiency. While it is possible that mineral composition could affect solubility or stabilization of certain phytochemicals, this mechanism was not directly investigated. Previous reports have described the unique mineral profile of Zamzam water17,34, but its role in phytochemical extraction remains insufficiently characterized. Therefore, the enhanced activity associated with Zamzam extracts should be interpreted cautiously.

Resistance patterns were also observed. Certain aqueous and petroleum ether extracts showed limited or no inhibition against specific organisms. These findings may reflect intrinsic resistance mechanisms, including reduced membrane permeability or efflux systems commonly reported in Gram-negative bacteria33. However, no molecular resistance assays were conducted in the present study; thus, mechanistic explanations remain inferential.

Molecular docking analysis identified several phytocompounds with favorable predicted binding affinities toward DNA gyrase and dihydrofolate reductase. β-Caryophyllene, quercetin derivatives, and related flavonoids demonstrated strong predicted interactions within active-site regions. Previous studies have reported antimicrobial and enzyme-inhibitory properties of flavonoids and terpenoids4,31. However, docking scores represent theoretical binding estimations and do not confirm biological activity or mechanism. No direct statistical correlation between docking scores and inhibition zone diameters was performed.

Several limitations should be considered when interpreting these findings. The antimicrobial assays were conducted in vitro and do not account for pharmacokinetic or host-related factors. Crude extracts rather than isolated compounds were evaluated, limiting attribution of activity to specific phytochemicals. Additionally, total phenolic or flavonoid content was not quantified, and potential synergistic interactions among extract constituents were not investigated.

Conclusion

This study demonstrates that solvent type substantially influences the antimicrobial activity of the investigated plant species. Across the tested microorganisms, ethanol and methanol extracts consistently produced larger inhibition zones, while Zamzam water extracts showed activity comparable to alcohol-based extracts in several cases and higher activity than distilled water extracts. Among the evaluated plants, C. speciosus and L. usitatissimum exhibited the most consistent broad-spectrum inhibition patterns.

Molecular docking analyses identified several phytocompounds with favorable predicted binding affinities toward DNA gyrase and dihydrofolate reductase, providing mechanistic insights that may partially support the observed in vitro antimicrobial effects. However, these computational findings represent predictive models and do not confirm biological efficacy.

Given the in vitro nature of this study, the results should be interpreted as preliminary. Future research should prioritize (i) isolation and quantitative characterization of the most active fractions, (ii) evaluation of dose–response relationships and minimum inhibitory parameters under standardized conditions, and (iii) in vivo validation to assess pharmacological relevance and safety. Such stepwise investigation will be essential before considering translational or therapeutic applications.