Biological assessment of Coccinia grandis leaf and Lupeol against β-lactam resistant Klebsiella pneumoniae through integrated in-silico and in-vitro studies

Abstract

The current study investigated a detailed account of phytocompounds within Coccinia grandis using GC-MS coupled with high-performance liquid chromatography. An in-silico approach was employed to gain insight into the inhibitory mechanism of Lupeol on Klebsiella pneumoniae. The molecular dynamics simulations were conducted over 100 ns to investigate the stability of metallo-β-lactamase with Lupeol, Imipenem (IPM), and Meropenem (MRP). We meticulously explored the antimicrobial activities of the crude extract of C. grandis leaves (CGL) and Lupeol against carbapenem-resistant K. pneumoniae. Additionally, cellular disruption activity was verified using a scanning electron microscope to better understand the antimicrobial activity of Lupeol. The computed free binding energy evaluated for Lupeol was − 92.380 ± 2.261 kJ/mol. This substantial negative value suggested the robust inhibitory potential of Lupeol, indicating a strong and stable interaction between Lupeol and target protein gold-bound NDM-1. Furthermore, the in-vitro antimicrobial activity of CGL and Lupeol were compared with standard antibiotics; MRP and IPM through the disc diffusion method. The zone of inhibition, and minimum inhibitory concentration, were found to be 23 ± 0.57 and 0.02 ± 0.01 mg/ml for Lupeol, 14.33 ± 0.58 mm and 0.03 ± 0.01 mg/ml for CGL. The ZOI of 12.67 ± 0.58 mm for IPM and 12.33 ± 0.58 mm for MRP was observed whereas 0.13 ± 0.06 mg/ml and 0.17 ± 0.06 mg/ml of MIC was determined for IPM and MRP respectively. Furthermore, the mechanism of action of Lupeol demonstrated significant activity in cell wall disruption assay and NDM-1 enzyme inhibition assay. Moreover, Lupeol also showed apoptotic activity against various cancer cell lines and no cytotoxic effect against healthy cell line. This study suggested Lupeol as an alternative natural therapeutic compound as an inhibitor of β-lactam resistant K. pneumoniae.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-41907-3.

Introduction

Medicinal plants are safer to utilize frequently than conventional antibiotics as 80% of plants were earlier proven as non-toxic¹. Because they are abundant in active phytochemicals with antimicrobial action, novel antibacterial medications derived from plants have emerged in recent decades^2,3. In the traditional system, Coccinia grandis has been reported as an ethnomedicinal plant with a wide range of pharmacological properties and is used for culinary purposes⁴. Treatments for diabetes, leprosy, jaundice, asthma, bronchitis, skin eruptions, burns, tongue sores, earache, indigestion, eye infections, nausea, insect bites, and fever are among its many common uses⁵. Additionally, some available studies have reported the antibacterial, antioxidant, and cell proliferative properties of different parts of C. grandis^6,7. Certainly, C. grandis leaf extract has been used as a therapeutic agent against severe ear infections⁸. In this respect, C. grandis is used in this study as an antibacterial therapeutic against deep neck infection bacteria of the head and neck region. It was noticed that deep neck infections in the head and neck region, along with the failure of antimicrobial treatment, mainly occur when the infection is secondary to dental caries, trauma, airway obstruction, oropharyngeal infection, severe ear infection, and neck abscess. However, the dangerous condition tends to get worse over time, especially when deep neck infections become the reason to cause of cancer⁹. Somehow, this infection increases the death rate by approximately 4.9% in adults and 6.2% in children¹⁰.

In 2017, the World Health Organization (WHO) gave global priority to guide the research and to discover new antibiotics for ESKAPE (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, Enterobacter species) pathogens. In this instance, carbapenem antibiotics are considered the first line of treatment and most used antibiotics for infections caused by the most resistant bacteria¹¹. With the growing frequency of antibiotic-resistant pathogens against clinically significant antimicrobials, empiric drugs and new generations of antibiotics are becoming resistant to infectious diseases¹². The International Guidelines for the Management of Sepsis (2016) suggest that carbapenems can replace the initial empiric antibiotics in cases where the illness becomes uncontrolled and evolves into sepsis without a known pathogen¹³. However, the bacteriological evaluation of patients presenting with deep neck space infection according to the severity and clinical variables found some strains of carbapenem resistance adhering to metagenomic sequencing¹⁴. In this situation, research into potential novel pharmaceuticals and antibacterial ingredients derived from a variety of natural sources has been ongoing^15,16.

Consequently, there have been incidences of toxicity and frequent clinical failure that have come across optimal carbapenem exposure¹⁷. Common adverse effects such as skin rash, digestive disorder, increased liver enzyme levels, increased creatinine/urea levels, and hematological disorders were observed¹⁸. Imipenem (IPM) and Meropenem (MRP) are the most approved antibiotics, where imipenem is highly resistant to β-lactamase enzymes produced by many multiple drug-resistant Gram-negative bacteria, thus recognized as a new generation of β-lactam antibiotic. In contrast, it was evident that this drug administration induced nephrotoxicity in animals and oxidative stress in plants¹⁹. However, MRP is mainly eliminated as it affects renal function. The acute renal disorder leads to a significant increase during the drug treatments, with a high risk of toxic adverse effects²⁰.

Significantly, plant-based drugs have become an urgent need for eluding the existing threat of drug resistance and the toxic effect of the drugs as well. In addition to that, molecular modeling, computational biology, and other synthetic chemical techniques have made great strides in drug discovery in recent years. Still, natural product-derived compounds remain an important source of pharmaceuticals for humans²¹.

The plant C. grandis has various secondary metabolites that are known to have a variety of beneficial properties, including flavonoids, phenolic acids, terpenoids, alkaloids, sterols, and glycosides²². Different solvent extracts using C. grandis plant parts suggested antimicrobial activity against different pathogens like Pseudomonas aeruginosa, Klebsiella pneumoniae, Bacillus licheniformis, Staphylococcus aureus, and Escherichia coli earlier^23,6. However, the phytocompound analysis is still needed to know the individual beneficial properties of the identified phytocompounds. Lupeol is a triterpene found in edible fruits, vegetables, and medicinal plants. A plethora of studies have demonstrated an extensive range of pharmacological activities such as anticancer, antioxidant, anti-inflammatory, and antimicrobial activities^15,24. The majority of non-steroidal anti-inflammatory Drugs (NASID) medications prescribed globally have serious side effects. In contrast to NSAID medications, natural medicine exhibits dual inhibition of the inflammatory and analgesic target proteins with greater effectiveness and fewer adverse effects^24,25.

Several studies have reported the antibacterial activity of Lupeol and Lupeol derivatives against Klebsiella pneumoniae isolates. It was evident that Lupeol was proven as an antibacterial agent against recurrent urinary tract infection in women, in management of biofilms on medical devices, and respiratory skin or infection^26,27. However, there were no such findings reported the antibacterial efficiency of Lupeol against carbapenem-resistant K. pneumoniae in deep neck infection.

In-silico approaches have proved beneficial for validating the unattributed biological activities of compounds and targets. Compared to the conventional drug discovery approach, computational screening appears to be a significant way to reduce time, money, and human resources. In addition, it has tested binding affinities to targets and predicted physicochemical qualities from molecular and structural aspects of many different chemical compounds^28,29.

The present study emphasizes the exploration of phytocompounds present in the C. grandis through GC-MS analysis. Subsequently, the mechanistic insight into the inhibition of K. pneumoniae by Lupeol was examined through an in-silico approach such as molecular docking as well as molecular dynamics simulations with the Metallo-β-lactamase type 2 protein. Moreover, the antimicrobial investigation of crude extract of C. grandis against carbapenem-resistant K. pneumoniae was explored. Subsequently, the phytocompound Lupeol identified from C. grandis was also assessed for antimicrobial investigations to inhibit carbapenem-resistant K. pneumoniae.

Results

Detection of the β-lactam-resistant gene

The target genes from K. pneumoniae were successfully amplified using gene-specific primers, with product sizes verified against a 100 bp DNA ladder (100–1500 bp, Gene to Protein Pvt. Ltd.). The specific amplification sizes for each target are detailed in Table S1 and visualized in Fig. 1. Following amplification, the PCR products were resolved via agarose gel electrophoresis and visualized under UV illumination.

Fig. 1.

The agarose gel electrophoresis results illustrate the PCR amplification of genes linked to β-lactam resistance in Klebsiella pneumoniae. Lane L contains a 100 bp DNA ladder ranging from 100 to 1500 bp (Gene to Protein Pvt. Ltd.). The lanes display the amplification of 16 S rRNA (serving as an internal control), blaOXA-23, blaOXA-51, blaSIM, ISAba, and blaNDM-1, each amplified using gene-specific primers. The presence of distinct bands at the anticipated amplicon sizes verifies the existence of the respective resistance genes.

To emphasize pertinent bands, the images presented in the main figures have been cropped for clarity. However, to ensure full transparency and continuity, the gels were maintained intact throughout the electrophoresis and imaging processes. Uncropped gel images of molecular weight markers and gel edges are provided in the Fig. S1.

GC-MS and LC-MS analysis

GC-MS and LC-MS analysis was used to identify biologically active phytochemicals such as phenolics, flavonoids, and triterpenes in the extract obtained from C. grandis leaves (CGL). GC-MS and LC-MS was performed on methanolic extracts of CGL and recorded a total of 23 peaks from GC-MS analysis and 35 peaks from LC-MS analysis. The identified GC-MS analysis bioactive compounds are categorized by relating their peak retention time, peak area (%), height (%), and mass spectral fragmentation pattern to that of the known compounds. Whereas for LC-MS analysis the identified bioactive compounds were distinguished with retention time (min), molecular formula, mass (m/z), mass (Tgt), diff (Tgt, ppm) and obtained score of mass spectral fragmentation compounds. The GC-MS chromatogram of the methanolic CGL extract is presented in (Fig. 2) along with their peak area and retention time.

Fig. 2.

The chromatogram of the methanolic C. grandis leaf extract was by using GC-MS and quantifications were done by mass spectrometry coupled with Ultimate 3000 high-performance liquid chromatography.

The primary 23 recorded compounds in CGL were mentioned in the supplementary Table S2, with their retention time, molecular formula, molecular weight, peak area, and structure. Whereas, the 35 identified compounds from LC-MS analysis were mentioned in supplementary Table S3. Among all, Lupeol was recorded with the highest retention time of 50.133 min, compared to all phytocompounds from GC-MS analysis. Whereas, it was also observed Lupeol at retention time of 24.674 min and represents identical score of 96.53% from CGE extract from LC-MS analysis. Although, the phytocompounds identified by GC-MS analysis were categorized under some biologically active groups such as phenols, flavonoids, terpenoids and triterpenoids, alkaloids, sterols, glycosides, and more, which were known to have various health-beneficial activities.

Molecular docking

In accordance with GC-MS analysis, we have further performed molecular docking studies for all 23 compounds identified in the Coccinia grandis extract. The comprehensive binding energy data, interaction profiles, and docking scores for these compounds have been compiled in the Supplementary Table S4. With previewing of docking score and the most abundant bioactive compound Lupeol, MRP, and IPM was selected for molecular docking simulations insights to examine the interactions with the β-lactamase type-2 target. During molecular docking, we found some common amino acids that participated in interactions with the Lupeol and the control molecules IPM and MRM. The Lupeol exhibited interactions with the ASP124 through hydrogen bonding represented by green dotted lines; similar interactions were shown by the MRP and IPM with the binding site of the metal β-lactamase type-2 target, Fig. 3 (A-F). The alkyl and pi-alkyl contacts shown by the Lupeol, MRP, and IPM are represented by pink dotted lines. The binding score generated by the Lupeol was found to be -6.6 kcal/mol, whereas binding energy exhibited by the IPM and MRP was found to be -5.6 and − 6.6 kcal/mol. The more intentions to the Lupeol because the retention time was higher than that of other compounds confirmed by GC-MS analysis, so we therefore reported the binding analysis in detail for Lupeol, and control molecules IMP, and MRP. Moreover, the bond distance shown by the Lupeol with ASP142 of 6LHE is 1.86 Å. The bond distances observed between the IPM and 6LHE binding site amino acids ASP124, GLU152, ASP223, and HIS250 were found to be 1.45, 1.40, 2.51, 2.87, and 2.68 Å, and the bond distances observed between the MRP and ASP124, HIS189, and HIS250 were found to be 1.31, 2.60, and 2.50 Å, respectively. These results suggest that each control molecule, as well as the Lupeol, showed better bindings and accommodated in the binding site of β-lactamase type-2 target.

Fig. 3.

The molecular docking interactions observed among Lupeol, IMP, and MRP with 6LHE, the hydrogen bonds (donor-acceptor), hydrophilic and hydrophobic contacts, and polar contacts are represented as blue dashes, grey, and cyan. (A, B) The gold bound NDM-1 (PDB ID: 6LHE) complexed with Lupeol and 3D binding interactions. (C, D). The 6LHE complexed with IPM, and 3D binding interactions (E, F). The 6LHE complexed with MRP, 3D binding interactions in the binding site. The upper panel shows the 6LHE bound with Lupeol, IMP, and MRP molecules, and the lower panel depicts the molecular interactions with the amino acids in the active site.

Molecular dynamics simulations

We employed a molecular dynamics simulation approach to verify the stability and conformational changes of the ligand taken in this study. The RMSD is a useful technique to measure the structural stability of a system over time by comparing the deviation of the current frame from the initial reference frame. It provides insights into conformational changes, equilibration, and the overall structural drift of the protein or ligand during the simulation. We also aimed to examine the conformational changes in the β-lactamase type-2 target, gold bound NDM-1 (PDB ID: 6LHE) during simulations. The Lupeol conformations were recorded on a nanometer scale and conformational changes were plotted in the form of root mean square deviations (RMSD). Initially, the gold bound NDM-1 (PDB ID: 6LHE) complexed with the Lupeol showed higher conformational till 20 ns of the simulation period, and the RMSD was found to be 0.5–1 nm then the Lupeol conformations were found to be in a declined trend, and stability was achieved by in the binding site of the gold bound NDM-1 (PDB ID: 6LHE) target. The Lupeol conformations were found linear and the RMSD was found to be 0.4–0.7 nm till the end of the simulation period. The Lupeol trajectory is represented in the green color. Also, the IPM and MRP showed stability parallel to the Lupeol for a 100 ns period and the RMSD ranged between 0.35 and 0.7 nm (Fig. 4A). The conformations of the ligands such as (Lupeol, IPM, and MRP) were also determined after superposing into the respective ligands and the RMSD trajectories obtained were ranging between 0.02 and 0.3 nm (Fig. 4B). This result suggests that Lupeol is stable in the binding site of the 6LHE and showed comparable stability followed by the IPM and MRP.

Fig. 4.

(A) The stability of the ligands was observed upon the superposing of the 6LHE protein for 100 ns of the simulations in an aqueous medium, the colour given to each ligand is mentioned in the legend box of the respective plot; (B) The stability of the only ligands was observed upon the superposing on the respective ligand for 100 ns of the simulations in an aqueous medium; (C) The stability of the whole target protein 6LHE was observed upon the superposing backbone chain to protein for 100 ns of the simulations in an aqueous medium.

Next, the whole protein (6LHE) conformations were investigated for a long-range simulation run. The three complexes were put in a record in terms of stability, the lower the occupancy of the conformations during simulations, the higher the stability of the protein. The RMSD of backbone Cɑ of each complex was calculated and the trajectories obtained were analyzed. The calculated RMSD of the 6LHE complexed with Lupeol reaches equilibrium at 32 ns of simulation period then the conformations will remain linear. The RMSD of 6LHE complexed with Lupeol was found to range between 0.2 and 0.3 nm (Fig. 4C). These values reflect that the 6LHE complexed with Lupeol was stable and no major conformational changes were observed. Next, the 6LHE complexed with IPM and MRP showed similar conformational changes during simulations. However, the backbone Cɑ results suggest the 6LHE complexed with Lupeol archives parallel stability with 6LHE complexed with IPM and is more stable than the 6LHE complexed with MRP.

The RMSF evaluates the flexibility of individual residues or atoms in the protein, highlighting regions with significant motions or movements. The flexible regions often correspond to active or binding sites, loops, or regions involved in dynamic interactions, while stable regions suggest structural rigidity. Higher flexibility leads to structural instability. The root mean square fluctuations for each complex were monitored in 6LHE complexes with Lupeol, IPM, and MRP. The aim of determining the root mean fluctuations in each amino acid was to observe the effects of the protein upon ligands ligand bindings. The N-terminus did not show higher fluctuations; these fluctuations are found in the permissible limits whereas the higher fluctuations were observed at C-terminus. Moreover, most of the amino acids in each complex showed fluctuations of 0.05–0.25 nm on average but three positions were observed where the fluctuations were higher and those positions ranged between 125 and 125, 145–155, and 215–222. However, the binding site does not show any effect on the bindings of Lupeol, IPM, and MRP (Fig. 5A).

Fig. 5.

(A) The root mean square fluctuations of each amino acid present in the protein 6LHE were determined for 100 ns of the simulations; (B) The compactness of each complex was calculated from the data points obtained from the radius of gyrations for 100 ns of the simulation period; (C) Hydrogen bonds formed during MD Simulations for 100 ns by Lupeol, IPM, and MRP; (D) Energy contributed by each amino acid during free energy calculations using the MM-PBSA method for phytocompound Lupeol and control drug molecules MRP, and IPM in kJ/mol.

The radius of gyration (RoG) assesses the compactness of a molecule by calculating the distribution of atoms around their center of mass. A declining or stable RoG indicates folding, compactness, or retention of structural integrity, which is particularly important for determining protein stability and folding dynamics. The radius of gyration is an essential parameter in molecular dynamics simulations to measure the compactness of the complexes. The compactness is measured upon the linearity of the RoG (radius of gyration) values. The RoG value of a protein is subject to variations stemming from conformational alterations and structural dynamics. When the RoG value is elevated, it signifies that the protein is in a more expanded or flexible state. Conversely, lower RoG values imply a more compact or rigid conformation of the protein. These fluctuations in RoG offer insights into the inherent flexibility and structural characteristics of the protein, reflecting its ability to undergo conformational changes. All three complexes showed linearity in the RoG values therefore each complex is stable and less flexible (Fig. 5B).

The number and persistence of hydrogen bonds are critical indicators of molecular interactions, such as ligand-protein binding or intramolecular stability. Consistent hydrogen bonding correlates with strong, stable interactions, whereas variations can signal dynamic binding events or instability. This analytical approach proves valuable for studying molecular dynamics and system behaviour under various conditions, providing a nuanced understanding of how molecules interact and systems evolve. During molecular dynamics simulations the Lupeol, IPM, and MRP forms 4, 6, and 11 hydrogen bonds contacts with the 6LHE target protein (Fig. 5C and D). This current study demonstrated that each ligand is stable in the binding site of the 6LHE target protein; also, a single stable hydrogen bond may reflect the stability of the ligands.

Conformational analysis

The conformational analysis during simulations of each complex was performed by isolated complexes with the 10 ns time interval. The prime focus was to visualize the conformational changes occurring in protein as well as ligands complexed with the protein. Those conformations provide depth insight into the changes occurring in the complexes during simulations The isolated complexes were aligned with each other and the RMSD was calculated; the total RMSD of the 6LHE complexed with (A) MRP, (B) IPM (B) Lupeol showed RMSD 2.3220, 2.2185 and 1.7165 Å. From the conformational investigations the RMSD of the aligned complexes was found with permissible limits and the RMSD of Lupeol complexed with 6LHE was found minimum than the other two reference complexes Overall, Lupeol has a potency to remain intact with the target protein and therefore inhibit the function of the β-lactamase therapeutic target of K. pneumoniae (Fig. 6).

Fig. 6.

The conformational analysis was performed to investigate the changes occurring during the molecular dynamic simulations for 6LHE complexed (A) MRP, (B) IPM, and (C) Lupeol after every 10 ns of the time interval.

Principal component analysis (PCA), Gibbs free energy landscape and secondary structure analysis

Protein flexibility was assessed using essential dynamics, with Principal Component Analysis (PCA) applied to study the motions of β-lactamase complexed with Lupeol, IPM, and MRP. Eigenvalues and eigenvectors (PC1, PC2) were derived using gmx covar and gmx angle. The Gibbs energy landscape, color-coded (blue: maxima, red: minima), revealed conformational stability. The Lupeol-bound β-lactamase showed the most stable state, comparable to MRP and IPM complexes (Figs. 6 and 7). Meta-stable conformations were identified using gmx sham and gmx xpm2ps. The MRP and Lupeol complexes exhibited superior stability over IPM (Fig. 8), with Lupeol displaying meta-stable conformation to reference-bound complexes. Free energy surface (FES) values were 18.4 kJ/mol (IPM), 18.1 kJ/mol (MRP), and 18.1 kJ/mol (Lupeol), indicating similar stability. These findings suggest Lupeol may effectively inhibit β-lactamase.

Fig. 7.

The principal component analysis was performed using gmx aneig and gmx eigen modules. The most compact structure obtained from the PCA is β-lactamase bound Lupeol colored in green.

Fig. 8.

The Free energy surfaces of the MRP, IPM, and Lupeol were investigated from the 100 ns trajectory. The blue colour represents most stable conformations, green represents average stable conformations and yellow represents less stable conformations. (A) β-lactamase bound IPM, (B) β-lactamase bound MRP and (C) β-lactamase bound Lupeol. The most stable structures were represented in pink colour.

Furthermore, we also analyzed three 6LHE complexes (bound to MRP, IPM, and Lupeol) over 100 ns using DSSP to assess structural stability and transitions. Comparing initial (0 ns) and final (100 ns) states, we observed ligand-specific variations in β-sheets, helices, turns, and other elements (Fig. 9 and Table S5). The results highlight the protein’s adaptive response to different ligands, enhancing understanding of ligand-protein interaction dynamics.

Fig. 9.

The rainbow representation of the secondary structures of the gold-bound NDM-1 (PDB ID: 6LHE) β-lactamase complexed with (A) MRP, (B) IPM, and (C) Lupeol. *A- Alpha, B- beta.

Post docking interactions analysis

After MD simulations, the protein-ligand complexes were isolated at the end of simulations (100 ns) and were processed for molecular interactions using DS studio software. The interactions exhibited by the Lupeol with amino acids Val73 and Leu 221 showed pi-alkyl contacts with Lupeol. Leu 65 and Trp93 form alkyl contacts, and Asp124 showed a hydrogen bond contact as well as carbon hydrogen contact with Lupeol. Further interactions were assessed with the 6LHE complexed with IPM, and contacts observed were Gly71, Gln123, Asp 124, and Glu 152 form conventional hydrogen bond whereas Glu65, Val73, and Trp93 showed pi-alkyl/alkyl contacts with IPM. Furthermore, the 6LHE complexed with MRP showed carbon-hydrogen contacts with amino acids (Met67, His120, His122, Lys125, and His189). The pi-alkyl/alkyl contacts were observed with amino acid (Val73, Trp93, and His250). The complexes are represented in Fig. 10A, C and D, Moreover, the hydrogen bond contact was observed with amino acid Asp124 (Fig. 10B, D, E, F). In post docking analysis, it was noticed that Asp124 is a common amino acid that interacts with all three ligands selected in the present study. In molecular docking as well as the post interactions analysis the Asp124 was found to be a prime amino acid that interact with Lupeol, IPM and MRP. This offers a basic foundation for the development of novel natural products to inhibit K. pneumoniae.

Fig. 10.

(A, B) Post-docking interactions of the Lupeol with 6LHE, (C, D) Post-docking interactions of the IPM with 6LHE, (E, F) Post-docking interactions of the MRP with 6LHE.

Free binding energy calculations

The binding free energy exhibited by Lupeol was − 92.380 kJ/mol, whereas MRP and IPM showed more negative binding energy than that of Lupeol, as concluded from this study. The binding free energy calculated from the MM-PBSA method demonstrated that Lupeol was also a potent target of the Metallo-β-lactamase type 2 therapeutic target of K. pneumoniae (Table 1). During the MRP free binding energy calculation, we assessed the contributions of energy exhibited by each amino acid. We generally consider top contributed free energy amino acids, which exhibited energy of more than − 3 kJ/mol, are active, and active amino acids are LEU-65 -4.36, MET-67 -4.4, VAL-73 -3.12, TRP-93 -10.47, HIS-120 -4.70, and LYS-125 -12.18 kJ/mol. Next, the IPM upon ΔG_binding calculations, the following amino acids contribute to free energy are MET-67 -3.43, GLY-71 -3.58, VAL-73 -5.89, TRP-93 -4.67, GLN-123 -3.46, ASP-124 -22.04, MET-126 -3.42, ASN-220 -4.76, and HIS-250 -5.81 kJ/mol. Moreover, the free energy contribution by the amino acids upon 6LHE complexed with Lupeol was LEU-65 -6.83, VAL-73 -5.21, TRP-93 -3.65, and LEU-221 -6.79 kJ/mol. The free energy contribution demonstrates that the above-mentioned amino acids are active and participate in the energy contribution.

Table 1.

The free energy bindings profile of Lupeol, MRP, and IPM was calculated by the MM-PBSA method.

Compounds	ΔEvdw (kJ/mol)	ΔEElec (kJ/mol)	ΔGpolar (kJ/mol)	SASA (kJ/mol)	ΔGbinding (kJ/mol)
Lupeol	− 180. 3 ± 2.8	− 2.022 ± 1.6	29.3 ± 1.7	− 11.3 ± 0.4	− 92.4 ± 2.3
Meropenem (MRP)	− 152.5 ± 2.1	− 60.080 ± 3.7	84.4 ± 1.9	− 15.1 ± 0.2	− 143.1 ± 2.9
Imipenem (IPM)	− 85.8 ± 4.6	− 242.030 ± 14	150.7 ± 10.2	− 11.4 ± 0.4	− 189.9 ± 6.8

ADMET analysis

The ADMET analysis was performed for compounds MRP, IPM and Lupeol (Table 2). This analysis suggested that the MRP and IMP are hepatotoxic compounds with lower affinity to cross the blood brain barrier and central nervous system. Whereas the Lupeol did not show hepatotoxicity and AMES toxicity and cross the blood brain barrier. The MRP, IPM and Lupeol does not inhibit the normal function of the CYP isoforms. Moreover, the molecular properties showed all molecules that have MW > 500 therefore could easily pass through the membrane protein channels such as voltage dependent anion/small channel proteins and meet the biological target. The water solubility of all three molecules was very low, and are much readily soluble in octanol over water. The intestinal absorption of MRP, IMP is very low than the Lupeol. The MRP, and IMP is not a glycoprotein inhibitor but the Lupeol inhibits the glycoprotein, this reflects that the Lupeol may inhibit multitargets of the bacterial stains.

Table 2.

ADMET analysis performed for compounds MRP, IPM and Lupeol.

Molecule properties		Absorption		Digestion		Metabolism		Excretion		Toxicity
Meropenem
Molecular Weight	383.4	Water solubility	− 2.568	VDss	− 1.191	CYP2D6 substrate	No	Total Clearance	0.365	AMES toxicity	No
LogP	− 0.3	Caco2 permeability	0.028	Fraction unbound	0.766	CYP3A4 substrate	No	Renal OCT2 substrate	No	Max. tolerated dose (human)	0.793
Rotatable Bonds	5	Intestinal absorption	34.839	BBB permeability	− 0.754	CYP1A2 inhibitor	No			hERG I inhibitor	No
H bond Acceptors	6	Skin Permeability	− 2.735	CNS permeability	− 3.705	CYP2C19 inhibitor	No			Oral Rat Acute Toxicity (LD50)	1.815
H bond Donors	3	P-glycoprotein substrate	Yes			CYP2C9 inhibitor	No			Oral Rat Chronic Toxicity (LOAEL	2.17
		P-glycoprotein I inhibitor	No			CYP2D6 inhibitor	No			Hepatotoxicity	Yes
		P-glycoprotein II inhibitor	No			CYP3A4 inhibitor	No			Skin Sensitisation	No
										T.Pyriformis toxicity	0.285
										Minnow toxicity	3.695
										Oral Rat Acute Toxicity (LD50)	1.815
Imipenem
Molecular Weight	299.352	Water solubility	− 1.704	VDss	− 1.292	CYP2D6 substrate	No	Total Clearance	0.749	AMES toxicity	No
LogP	− 0.3857	Caco2 permeability	− 0.201	Fraction unbound	0.668	CYP3A4 substrate	No	Renal OCT2 substrate	No	Max. tolerated dose (human)	1.399
Rotatable Bonds	6	Intestinal absorption	37.445	BBB permeability	− 1.145	CYP1A2 inhibitor	No			hERG I inhibitor	No
H bond Acceptors	5	Skin Permeability	− 2.735	CNS permeability	− 4.039	CYP2C19 inhibitor	No			Oral Rat Acute Toxicity (LD50)	1.785
H bond Donors	3	P-glycoprotein substrate	No			CYP2C9 inhibitor	No			Oral Rat Chronic Toxicity (LOAEL	1.167
		P-glycoprotein I inhibitor	No			CYP2D6 inhibitor	No			Hepatotoxicity	Yes
		P-glycoprotein II inhibitor	No			CYP3A4 inhibitor	No			Skin Sensitisation	No
										T.Pyriformis toxicity	0.281
										Minnow toxicity	3.263
Lupeol
Molecular Weight	426.729	Water solubility	− 5.861	VDss	0	CYP2D6 substrate	No	Total Clearance	0.153	AMES toxicity	No
LogP	8.0248	Caco2 permeability	1.226	Fraction unbound	0	CYP3A4 substrate	Yes	Renal OCT2 substrate	No	Max. tolerated dose (human)	− 0.502
Rotatable Bonds	1	Intestinal absorption	95.782	BBB permeability	0.726	CYP1A2 inhibitor	No			hERG I inhibitor	No
H bond Acceptors	1	Skin Permeability	− 2.744	CNS permeability	− 1.714	CYP2C19 inhibitor	No			Oral Rat Acute Toxicity (LD50)	2.563
H bond Donors	1	P-glycoprotein substrate	No			CYP2C9 inhibitor	No			Oral Rat Chronic Toxicity (LOAEL	0.89
		P-glycoprotein I inhibitor	Yes			CYP2D6 inhibitor	No			Hepatotoxicity	No
		P-glycoprotein II inhibitor	Yes			CYP3A4 inhibitor	No			Skin Sensitisation	No
										T.Pyriformis toxicity	0.316
										Minnow toxicity	− 1.696

The drug likeness and drug-like parameters were calculated using swiss adme accessed on 4-02-2025. During investigations it was found that all molecules (MRP, IMP and Lupeol) accept the Lipinski rule of five. The MRP and Lupeol accepts Veber rule, single violations are reflected by Egan rule and no violations were reported in Lupeol from Muegge. Next the Pan Assay Interference structure (PAINS) were evaluated for each structure and no alerts were reported in Lupeol including control molecules (MRP and IMP). The structural alert (Brenk) concludes no alert for MRP and 2 alerts, were reported for isolated (imine 1 & 2) for IMP molecule and single alert isolated alkene were reported for Lupeol (Table 3).

Table 3.

Drug likeness and medicinal chemistry parameters were evaluated for MRP, IMP and Lupeol.

	Meropenem	Imipenem	Lupeol
Drug likeness
Lipinski	Yes	Yes	Yes
Ghose	No, 1 violation	No, 1 violation	No,3 violation
Veber	Yes	No; 1 violation	Yes
Egan	No; 1 violation	No; 1 violation	No; 1 violation
Muegge	No; 1 violation	Yes	No
Bioavailability Score	0.55	0.56	0.55
Medicinal Chemistry
PAINS	0 alert	0 alert	0 alert
Brenk	0 alert	2 alerts: imine_1, imine_2	1 alert: isolated_alkene

In-vitro antibacterial study

Considering the need to develop new antibacterial drugs from CGL extract, we have analyzed the antibacterial activity of Lupeol, which was determined as the most stable compound through the above in-silico approach. Following the molecular docking and molecular dynamic simulation, the in-vitro antibacterial study of Lupeol can conclude with the efficacy of its inhibitory action against K. pneumoniae. However, developing the plant’s most stable and prominent compound is the best way to drug development. Lupeol was tested for its antibacterial activity using the agar-well diffusion method along with the crude and the antibiotics (IMP and MRP). The Lupeol exhibited inhibition of the 22.67 ± 0.57 mm (Mean ± Standard deviation) zone against K. pneumoniae, which was the moderate zone of inhibition compared to the crude (14.33 ± 0.58 mm), IMP (12.67 ± 0.58 mm) and MRP (12.33 ± 0.58). Moreover, the MIC was evaluated concerning the positive control antibiotic (Colistin) presented in Fig. 11.

Fig. 11.

Antibacterial activity of C. grandis crude extract and Lupeol (A) The zone of inhibition (ZOI) exhibited by the CGL, IPM, MRP, and Lupeol against K. pneumoniae, (B) MIC value of Lupeol, CGL, IPM, MRP, denoted by red squares; *NC-Negative control, CL-Colistin (positive control).

Determination of the particular concentration of bacterial inhibition was attributed through MIC, and MBC analysis by the comparative analysis with the crude and Lupeol at the same concentration of drug treatment in triplicate. The MIC and MBC analysis of Lupeol against K. pneumoniae was determined within 0.4-0.012 mg/ml. Significantly, the MIC of Lupeol was determined with 0.02 ± 0.01 mg/ml and MBC, 0.04 ± 0.01 mg/ml (Table S6). The MBC of Lupeol against K. pneumoniae was seen by subculturing K. pneumoniae inoculum from each well of the micro-titre plate on the nutrient agar plate. No microbial growth on the specific growth media was identified.

Visualisation of cell wall rupture activity on FE-SEM

As seen in Fig. 12, the micromorphology of K. pneumoniae was seen through the interaction of Lupeol with the cell surface. The bacterial cell walls were non-specifically adhered to the Lupeol solution, which caused cell wall rupture. Lupeol’s effectiveness against the β-lactam resistant pathogen K. pneumoniae was validated by the antibacterial effects visualized by FE-SEM.

Fig. 12.

Visualization of cell wall rupture activity of Lupeol against K. pneumoniae (A) Without treatment with Lupeol (B) Treated with Lupeol.

NDM-1 enzyme activity of Lupeol

The inhibitory effect of Lupeol phytocompound was further evaluated on NDM-1 enzyme activity. The result showed that both Lupeol and EDTA have significant NDM-1 inhibition activity (Fig. 13). The dose-response analysis revealed that EDTA and Lupeol inhibited NDM-1 at an IC₅₀ value of 0.000384 mg/ml and 0.001661 mg/ml respectively. These finding suggest that Lupeol exerts as inhibitor for NDM-1 though with slightly lower efficacy compared to EDTA.

Fig. 13.

(A) The inhibitory effect of Lupeol on NDM-1 enzyme activity. EDTA served as a positive control in these experiments. Error bars represent the mean ± standard deviation (SD) of three independent replicates. (B) The inhibitory effects of Lupeol, Lupeol + ZnCl₂, EDTA, and EDTA + ZnCl₂ on zinc-supplemented chelation ion NDM-1 enzyme activity. Statistical significance **** denoted as P < 0.0001.

Inhibitory activity of Lupeol and NDM-1 enzyme through zinc ion chelation

Zinc (Zn²⁺) ions are known to play a vital role in maintaining the hydrolytic activity of NDM enzymes, as evidenced by the inactivation of NDM-1 upon treatment with EDTA⁵⁷. With this leading background, the activity of Lupeol inhibition on NDM-1 by chelating Zn²⁺ using a zinc homeostasis assay. Based on zinc supplementary activity, Lupeol exhibited a moderate inhibitory percentage activity (49 ± 2) on the NDM-1 enzyme, whereas, the control EDTA shows mean percentage inhibition activity (30 ± 2) on the NDM-1 enzyme. Whereas, without zinc ions supplements both Lupeol and EDTA exhibits the hydrolysis rate 80% ± 2 and 87 ± 2.6 respectively on NDM-1. Overall, Lupeol demonstrates the hydrolysis activity was observed to be reduced by 50% (Fig. 13). These findings suggest that Lupeol partially exerts its inhibitory effect by chelating Zn²⁺ at the active site of NDM-1. Henceforth, the result of Lupeol treatment demonstrates an inhibitory activity by chelating zinc ions of NDM-1 hydrolytic activity.

Screening for anticancer potential and cytotoxicity

Lupeol was evaluated for their anticancer efficacy against cancer cell lines of different tissue origin (A549, OVCAR-3, MCF-7 and PC3) and HEK293 as non-cancer cell line. The cells were treated with Lupeol concentrations ranging from 0.012 to 0.4 mg/ml. The anticancer activity was assessed by monitoring cell proliferation and apoptosis induction in a concentration-dependent manner. Multiple assays were employed to confirm the cytotoxic effect of the extract. The IC₅₀ values were determined as 0.134 ± 0.017 (A549) mg/mL, 0.149 ± 0.013 mg/mL for OVCAR-3, 0.163 ± 0.028 mg/mL (MCF7) 0.123 ± 0.009 mg/mL (PC3) and ˂ 0.43 ± 0.03 mg/mL for HEK293 cells. The cytotoxic response was both concentration- and time-dependent, as depicted in Fig. 14.

Fig. 14.

Cell viability and apoptosis activity of Lupeol against (A) Cancer cell lines (A549, OVCAR3, MCF-7, PC3) (B) Non cancer cells Human embryonic kidney cell line (HEK-293); (C) Image shows Apoptosis assay by AO, EtBr treated with Lupeol and Adriamycin, each experiment has been repeated three times and SD has been measured, bar diagram shows relative % Apoptotic cells per focus, by AO, EtBr, staining.

Apoptosis assay by AO/EtBr staining with A549 cells treated with lupeol

Acridine Orange/Ethidium Bromide staining of A549 cells treated with Lupeol (0.4 mg/mL) and standard drug Adriamycine (0.27 mg/mL) demonstrated characteristic apoptotic features, including nuclear condensation and membrane disruption, as evidenced by green and orange/red fluorescence, respectively. The apoptosis activity of A549 cells treated by Lupeol, Adriamycin and untreated cells shows as 22 ± 0.3, 32 ± 1.4 and 5 ± 1.6 respectively. These findings suggest that Lupeol effectively induces apoptosis in A549 cells Fig. 14 and Fig. 15.

Fig. 15.

Acridine Orange and Ethidium Bromide staining of A549 cells treated with Lupeol and standard drug Adriamycine.

Discussions

The β-lactam-resistant strain K. pneumoniae (OP889580), used in this study was isolated from the head and neck region and identified through 16 S rRNA sequencing. The multidrug-resistant pathogen K. pneumoniae has been becoming more prevalent during the past 20 years as a result of its widespread distribution. Moreover, the rate at which carbapenem-resistant K. pneumoniae is developing is concerning²⁹. In this study, the β-lactam-resistant genes were detected using PCR amplification of the respective primers.

K. pneumoniae remained the most prevalent pathogen from most cases reported with deep neck infection of the head and neck region^30,31. Some studies presented it as a hypervirulent pathogen for its extensive drug resistance^32,33. Moreover, β-lactamase produced by antibiotic-resistant bacteria is another mechanism underlying the phenomenon of the evolution of multi-drug-resistant bacteria, and the clinical approaches to treat this resistant pathogen were insufficient and resulted in a markedly increased morbidity and mortality³⁴. Therefore, in this study, the C. grandis extract was proven as a key substance to find the biological therapeutics against K. pneumoniae using both in-silico and in-vitro approaches.

The phytocompounds analysed from methanolic CGL have already been reported from several plants^35,36 having different health benefits³⁷. Subsequently, the pharmaceutically important chemical ingredients found in C. grandis fruit methanolic and aqueous extract were alkaloids and phenolic compounds along with polyphenols, flavonoids, terpenoids, and triterpenoids⁵, which resembled the phytocompounds analyzed from methanolic extract of CGL in this study. The phytocompounds by GC-MS and LC-MS analysis of CGL may vary in different studies according to the geographical regions and solvents used. However, the phytocompounds analysed in this study were given in Table S2 and Table S3.

The phytocompound Lupeol identified in this study from methanolic CGL extract has the highest retention time. Various studies conducted with C. grandis have reported the presence of Lupeol as a major phytocompound with various inhibitory actions^38–40. However, the actual stability checks of the compound to combat the drug-resistant activity of bacteria is the benchmark of this study, as it was not defined earlier concerning both in-vitro and in-silico approaches. Thus, Lupeol was used as the antibacterial agent against the metal β-lactamase-positive K. pneumoniae isolated from head and neck infection.

Molecular dynamics simulations are a powerful tool to determine the movements of the atoms, and proteins in an atomistic scale⁴¹. In this study, we explore the stability of phytocompound Lupeol obtained from the C. grandis leaf extract followed by the GC-MS analysis and show comparative analysis with the control drug molecules IPM and MRP in terms of stability and bindings with the metal β-lactamase type-2 therapeutic target. The stability of each compound was observed from the conformational changes that occurred during the simulation period. The linear RMSD trajectory corroborates that the ligand is stable therefore, our results showed good agreement with several reports^42,43. The results of the conformational investigations showed that the RMSD of the aligned complexes was within acceptable limits, and the RMSD of the Lupeol complex with 6LHE was less than that of the other two reference complexes (IMP and MRP). Having been considered, Lupeol can bind to target protein and inhibit the action of K. pneumoniae’s β-lactamase therapeutic target. However, considering the toxicity effects^19,20 of these two reported antibiotics, Lupeol would give more justice in the case of alternative natural therapeutic approaches.

Hydrogen bond analysis serves as a standard practice in molecular dynamics simulations, allowing the monitoring of hydrogen bond formation within dynamic systems involving receptors and ligands⁴⁴. This technique involves establishing specific distance and angle criteria for hydrogen bond identification, typically defined by a predetermined cutoff distance between donor and acceptor atoms, as well as a minimum angle between the hydrogen bond vector and the acceptor-donor vector⁴⁵. Subsequently, the simulation tracks changes in distances and angles over time, identifying hydrogen bonds and offering insights into the strength and stability of intermolecular interactions.

The Principal Component Analysis (PCA), to assess protein flexibility within a simulated β-lactamase complex interacting with three distinct ligands: Lupeol, IPM, and MRP. A PCA scatter plot, based on Cα data, was instrumental in delineating the dynamic behaviors of the β-lactamase when complexed with each of the three ligands, thus contributing to a deeper understanding of the structural integrity and behavior of these protein-ligand complexes. Additionally, the Gibbs energy landscape for global maxima and global minima indicates energy states and thus, conformational energy stability. These methodologies, consistent with those used in prior studies, have proven valuable in analyzing protein conformation and meta-stable states during molecular dynamics simulations, thereby enhancing our comprehension of dynamics and stability^46,47.

Several previously reported studies^48,49 defined the antibacterial inhibition with crude extract of C. grandis. Based on earlier research⁵⁰, Lupeol’s inhibition activity exhibited by bacterial strain was 18 mm, whereas K. pneumoniae exhibited a better inhibition with Lupeol in this study. In contrast, most of the antibacterial studies with Lupeol^51,52 demonstrated antibacterial activity with higher concentrations, which defined the significance of this study.

Although with recent clinical observations for instance, a study of Klebsiella strains isolated from blood specimens reported that imipenem and meropenem isolates exhibited resistance with zone of inhibition diameter of 16 mm ⁵³. Similarly, in another study similar type of resistance pattern has been reported against β-lactam resistant K. pneumoniae⁵⁴. In our study demonstrated significant resistance to these carbapenems, with a Zone of Inhibition (ZOI) well below the CLSI susceptibility threshold, thereby validating the clinical relevance of our resistance model.

Furthermore, the antibacterial activity of Lupeol at the cellular level of K. pneumoniae was visualized in this study by FE-SEM analysis. Here the cell wall rupture activity of Lupeol demonstrated the possible therapeutic effect of Lupeol against the β-lactam group-resistant K. pneumoniae. It was evident in some previous studies that Lupeol promotes wound closure⁵⁵, and counteracts the proinflammatory signaling triggered in macrophages as therapeutics in atherosclerosis⁵⁶ but, studies with cellular validation are still necessary to fight against antibiotic resistance.

Zinc (Zn²⁺) ions are known to play a vital role in maintaining the hydrolytic activity of NDM enzymes, as evidenced by the inactivation of NDM-1 upon treatment with EDTA⁵⁷. The zinc supplementation assay confirmed that Lupeol inhibits NDM-1 by chelating its active-site zinc ions. Further, the mechanistic studies revealed that Lupeol exhibited potent inhibitory activity against NDM-1, with an IC₅₀ value of 0.001661 mg/ml. Based on these findings, it highlights that Lupeol is a promising lead phytocompound for combating carbapenem-resistant K. pneumoniae. With further optimization of Lupeol, the cytotoxicity was demonstrated with no toxicity effect against healthy cell lines and showing significant apoptotic activity as well. After being treated with Lupeol on various cancer cell lines the cell proliferation was monitored. Previously, the anticancer activity of Lupeol was suggested⁵⁸, but in response to carbapenem resistance K. pneumoniae, this study adds an in-depth analysis using both in-silico and in-vitro studies. However, it could be developed as a co-therapeutic agent alongside antibiotics to treat carbapenem-resistance.

Although in silico results showed higher binding affinities for imipenem and meropenem standard antibiotics and Lupeol demonstrated more robust in vitro inhibitory activity. This discrepancy because of predictive limitations of molecular docking, which primarily evaluates static binding at a specific pocket without accounting for pharmacokinetic factors or the dynamic physiological environment. Whereas in vitro assay, Lupeol shown superior membrane permeability and stability compared to chosen antibiotic, leading to higher effective concentrations at the target site. Furthermore, the target bacteria possess specialized resistance mechanisms such as efflux pumps and enzymatic degradation that significantly diminish antibiotic efficacy in vitro, a physiological complexity not captured by isolated protein-ligand docking.

Lupeol has antibacterial⁵⁹, antifungal⁶⁰, anticancer⁶¹, and anti-inflammatory^62–64, activity as well. Moreover, Liu et al. ⁵⁸ reported the molecular mechanism of Lupeol and its derivatives as anticancer and anti-inflammatory activity. However, no further study has been reported about the proper mechanism of antimicrobial activity. Research on some Lupeol derivatives is still in the preliminary stage; hence, more in-depth studies are essential for the future practical applications of these natural products.

Conclusion

The concluding remarks of the current study are that Lupeol showed good antimicrobial properties against carbapenem-resistant K. pneumoniae. Lupeol, a phytocompound, significantly reduces bacterial growth in antimicrobial screening against K. pneumoniae. Employing an in-silico approach, the inhibitory mechanism on K. pneumoniae reveals Lupeol displayed comparable binding affinity to standard antibiotic drugs; IPM, and MRP, targeting Metallo-β-lactamase type 2 (gold bound NDM-1). This in-vitro study demonstrated that CGL, Lupeol, IPM, and MRP inhibit the carbapenem-resistant K. pneumoniae followed by the experimental studies. Concerning the toxicity of both antibiotics, the in-silico and in-vitro studies demonstrated that Lupeol is not only a target of Metallo-β-lactamase protein as therapeutics but might have some additional targets through which the inhibition of K. pneumoniae occurred. Therefore, additional targets of the Lupeol will be addressed in subsequent studies. In addition to the above assay, in-vivo toxicity and genomic studies are required to elucidate the proper antimicrobial mechanisms of the Lupeol.

Materials and methods

Chemicals and reagents

The Lupeol (L5632) was procured from Sigma Aldrich and the NDM-1 protein (ABIN7479627) was purchased from Antibodies Online.

Collection of swab samples and bacterial culture analysis

A total number of 60 swab samples from head and neck infections were taken from various purulent discharge sites of patients admitted between 2020 and 2022 to the otorhinolaryngology department in a tertiary care teaching hospital. The scientific review committee and the Institutional Ethics Committee (IEC) approved this study (Ref. no/IEC/IMS.SH/SOA/2022/403), and the IEC follows the guidelines of the ICMR, Govt. of India. Informed written consent was obtained from subjects or legal guardians who provided the swab samples, and a thorough history regarding the patients’ primary symptoms, progression, and related symptoms was noted. The study was carried out in accordance with the Declaration of Helsinki followed guideline. Isolated swabs were processed further according to their standard procedures following Clinical and Laboratory Standards Institute (CLSI) guidelines and identified through VITEK-2 analysis. Furthermore, the identified K. pneumoniae samples was further analysed using universal 16 S rRNA primer⁶⁵. To confirm the β-lactam resistance, the gene-specific molecular characterization was carried out using carbapenem-resistant genes (Table S1). The carbapenem resistant primers OXA-23 (F- GATCGGATTGGAGAACCAGA, R- ATTTCTGACCGCATTTCCAT), OXA-51 (F- TAATGCTTTGATCGGCCTTG, R- TGGATTGCACTTCATCTTGG)⁶⁶, SIM (F- TTGCGGAAGAAGCCCAGCCAG, R- GCGTCTCCGATTTCACTGTGGC)⁶⁷, ISAba (F- CACGAATGCAGAAGTTG, R- CGACGAATACTATGACAC)⁶⁶, and NDM-1 (F- ATTAGCCGCTGCATTGAT, R- CATGTCGAGATAGGAAGTG)⁶⁴ were obtained from previous study and then procured from Barcode BioSciences Pvt.Ltd. (Bhubaneswar). In this study, K. pneumoniae (OP889580) was considered as an experimental strain maintained in glycerol stocks at -20 °C.

Detection of β-lactam-resistant genes

The β-lactam-resistant genes were detected using polymerase chain reaction (PCR) with the primers shown in supplementary Table 1 along with their annealing temperatures. The PCR was conducted with the amplification in a 25 µl reaction volume: 1×PCR buffer, 0.25 mM of dNTP mix, 10 picomoles of each primer, 0.25 U of Taq polymerase, and 50 ng/µl genomic DNA. The PCR reaction conditions were followed by initial denaturation at 94 °C for 4 min, then 35 cycles of denaturation for 30 s, annealing for 1 min, and extension for 2 min, again with a final extension at 72 °C for 7 min. The complete amplification process depends upon the annealing temperatures of the primers. After the amplification, the PCR products were verified using 1% agarose gel concerning 100 bp DNA Ladder (100–1500 bp) (Gene to protein Pvt. ltd.) and visualized in bioRad gel documentation.

Collection, identification, and preparation of plant material

The C. grandis leaves used in this experiment were collected from the Biju Patnaik Medicinal Plant Knowledge Centre, Bhubaneswar, Odisha. The required permission was obtained to collect the C. grandis plant. Then the plant was deposited at the Centre for Biotechnology, Siksha ‘O’ Anusandhan (Deemed to be University), Bhubaneswar, Odisha, India, and the plant specimen was verified by Dr. Pratap Chandra Panda (Voucher specimen No. 2548/CBT). At room temperature, 50 g of dried leaves were extracted using 95% methanol. The extract was concentrated and filtered in a rotary evaporator with reduced pressure. The dried sample was kept in a desiccator until further experiments. As described by Namchaiw et al., the C. grandis leaf (CGL) methanolic extract sample was diluted with 0.1% DMSO (dimethyl sulphoxide) and made different working concentrations of 400, 200, 100, and 50 mg/ml to use further⁴.

Phytocompound analysis by GC-MS and LC-MS/MS

The phytochemical evaluation of the methanolic CGL extract was initially performed on a GC-MS instrument. The Exactive TM Plus Orbitrap high-resolution mass spectrometer coupled with Ultimate 3000 high-performance liquid chromatography was used for qualitative and quantitative study of the phytochemical content of the methanolic extract (Thermo Scientific, USA). The extract (0.5 ml) and the 1:1 methanol: water solution (1.5 ml) were placed in a 2.0 ml centrifuge tube. The solution was centrifuged for 10 min at 4 °C and 10,000 rpm to remove any remaining solids after being properly mixed. A 0.22 M syringe filter was then used to purify the solution. To prepare for mass spectrometry, around 0.5 mL of the filtrate was aspirated and placed in a DP ID vial (Cat#C4000-1 W, Thermo Scientific, USA). Compounds in the solution were separated using a Hypersil BDS C18 (250 mm, 2.1 mm, 5 m; Thermo Scientific, USA) column. Methanol and water with 0.1% formic acid make up the mobile phase. The column temperature was kept at 30 °C, the pressure was kept at 700 bars, and a flow rate of 3 l/min was applied to the sample. The molecules were introduced to the orbitrap chamber through nitrogen gas and ionized using an electrospray ionization process at a voltage of 3 eV. After 15 min of analysis, both positive and negative ions in the 50–750 m/z scan range were discovered from the sample. Peak intensities were transformed into concentrations by comparing them to the standard curve, created by standardizing the peak intensities of the 5 standard compounds throughout the mass range of 50–500 m/z. After that, the mass peaks were annotated in a Python program using the PubChem library with an error range of 0.01 m/z. The phytocompounds were compared with the NIST library.

For LC-MS/MS analysis the following phytochemical profiling of methanolic CGL extract was analysed by using LC-ESI-QTOF-MS technology on an Agilent G6545XT Q-TOF system (Agilent Technologies, CA, USA). The chromatographic separation of methanolic extract (2 µl sample injection volume) was achieved on Agilent G7116A Column Compartment maintained at 35 °C. The mobile phases consisted of Water (solvent A), Methanol (solvent C), and ACN 90% (solvent D). The gradient elution program was as follows: 0–1 min, 100% A (0.3 ml/min); 1–6 min, A:C (100:0 to 65:35, v/v) (0.2 ml/min); 7–12 min, A:C (65:35 to 35:65, v/v); 13–20 min, A:C (35:65 to 0:100, v/v); 20–25 min, 100% C; and 25–29 min, 100% A. The total run time was 30 min with a stop time of 31 min. The flow rate was initially 0.3 ml/min and maintained at 0.2 ml/min throughout the gradient. Mass spectrometric detection was carried out in the m/z range of 50–1700 Da with a scan rate of 3 spectra/sec, using a Dual AJS ESI ion source in positive ion mode. The MS operating parameters were as follows: gas temperature, 325 °C ; nitrogen flow rate, 13 l/min ; nebulizer pressure, 40 psig ; sheath gas temperature, 285 °C ; sheath gas flow, 12 l/min ; nozzle voltage, 500 V ; capillary voltage (Vcap), 4000 V ; fragmentor potential, 175 V ; skimmer voltage, 65 V ; and Octopole RF Peak, 750 V. Additionally, UV-Vis signals were acquired using a DAD (model G7115A) at wavelengths of 220 nm, 280 nm, 250 nm, 320 nm, and 365 nm with a spectrum range of 190–400 nm. After execution of all steps the raw data was further processed by using MassHunter software (Agilent Technologies, CA, USA). The system was operated in MS1 acquisition mode with a scan range of m/z 50 to 1700 and a scan rate of 3 spectra/sec. The MS absolute threshold was set at 200 with a relative threshold of 0.010%. Chromatographic data were simultaneously recorded via a Diode Array Detector (DAD), with UV-Vis spectra acquired across a range of 190–400 nm with a 2 nm step. Specific signals were monitored at 220 nm, 280 nm, 250 nm, 320 nm, and 365 nm, all utilizing a 4 nm bandwidth. The identification of peak was achieved by comparing retention times and mass spectra against reference databases.

Ligand and receptor preparation

The phytocompound identified from the GC-MS analysis was sketched by the Avogadro software⁶⁹. The ligands were meticulously checked for the presence of hydrogen bonds, and other constituent atoms were added. Further, all phytochemicals, including MRP and IPM, were processed for minimization. The general amber force field (GAFF) was employed using the steepest descent algorithm, and the steps per update were selected to four. All chosen ligands were saved in the PDB format for further use. Moreover, the target protein used in this study was gold-bound NDM-1 (PDB ID: 6LHE), which was obtained from www.rcsb.org, as already evidenced by Sun et al. ⁶⁸. The Discovery Studio Software Version 21.1.0.20298 ⁷¹. (DS, 2021) was used to remove the water molecules and co-crystal ligands. The Metallo-β-lactamase type 2 therapeutic target of K. pneumoniae consists of two similar chains (A-B) and is composed of 242 amino acids in each chain. The chain A was retained, and the chain B was removed from the 3D X-ray crystallographic structure. The present investigations reveal that the starting amino acid of the Metallo-β-lactamase type 2 therapeutic target is GLY29 and ends with the amino acid ARG270. The protein preparation was followed by the previously reported protocols⁷², and the target protein was saved in the PDB format.

Molecular docking

The molecular docking was performed by the PyRx Software⁷³. The prepared protein (holo) was imported into the GUI of the PyRx Software and made into a macromolecule. The partial charges and hydrogen atoms were added to the x-ray crystallographic protein. Next, the open Babel module of PyRX was used to import the ligands (Lupeol, IPM, and MRP) into the graphical user interface for molecular docking simulations. The ligands were minimized again by using a default AMBER force field with the steepest descent algorithm and converted into the compatible Autodock pdbqt format. First, we identified the binding site using the PrankWeb server⁷⁴, and the binding site amino acids were A_122, A_123, A_124, A_152, A_189, A_208, A_211, A_215, A_217, A_219, A_220, A_223, A_250, A_251, A_67, A_73, A_93. Further, the grid box was built around the mentioned amino acids present in the binding site. The grid size of the binding site at center x, y, and z was 15.92, -2.06, and − 17.46 Angstroms. Also, the grid dimensions x, y, and z were 25, 16.16, and 16.57 Angstroms. The genetic algorithm was used for the docking search with default settings, and the exhaustiveness was set to 8. The top docking poses were selected and saved in the PDB format for further analysis. The interactions exhibited by the chosen ligands were observed by using the Discovery Studio softwarev21.1.0.20298 (DS, 2021)⁷¹.

Molecular dynamic simulations

The GROMACS 2022.4 software was employed to perform dynamic simulations of protein-ligand complexes⁷⁴. The 3D X-ray crystallographic structure of Metallo-β-lactamase type 2 (gold-bound NDM-1) (PDB ID:6LHE) was sourced from www.rcbs.org⁷⁰ and was docked following the above protocols. Both the docked ligand atoms and water molecules were eliminated from the three-dimensional structure. The AMBER99SB-ILDN.ff was implemented to allocate parameters for the protein^67,68 and the general Amber force field (GAFF) was utilized to generate the ligand topology by using the t-leap server^75,76. In addition, the prepared protein and ligand were merged into a solitary GROMACS-readable file. This file was positioned at the grid box’s center, 1 nm away from the periodic boundary conditions (PBC). The system was solvated by 12,667 solvents by using the gmx edit conf module⁷⁷. Further six Na+ ions were added to the prepared system and maintained the neutralization, with pH equivalent to ~ 7.4 met the biological system. Further, the steepest descent algorithm was utilized to perform energy minimizations, while the temperature and pressure were modified to 310 K and 1 bar, respectively⁷⁸. The Long-range electrostatic interactions are handled by the PME algorithm, whereas hydrogen bond lengths are handled by the LINCS algorithm⁷⁹. A 1.2 nm cut-off radius scheme was adhered to^80,81. Following that, the model system was equilibrated for a duration of 2 ns utilizing NVT and NPT ensembles⁸². The production run utilized a time step of 2 fs⁸³ to investigate the stability of ligands with the target protein; three complexes were simulated for a period of 100 ns. Furthermore, differential analysis various modules were used to analyze the dynamics data^84,85.We employed GROMACS modules to analyze the data such as RMSD, RMSF, RoG, H-bond, PCA, g_sham, and gmx xpm2ps.

Free energy binding calculation using the mm-pbsa method

Additionally, binding free energy calculations were conducted utilizing GROMACS version 5.1.4 ⁸⁶. The ensemble averaging of 100,000 snapshots extracted from a single, continuous 100 ns production trajectory was derived from molecular dynamics simulations employing the MM-PBSA method to compute the ΔG_bind⁸⁷. In addition to maintaining the ionic strength at 0.005 molar concentrations, the temperature was adjusted to 300 K. The free binding energy of Lupeol, IPM, and MRP with the metallo- β-lactamase type 2 therapeutic target of K. pneumoniae was determined using the subsequent equation.

1

The moments of the complexes during simulations were ultimately visualized using the Visual Molecular Dynamics (VMD) software^88–90.

ADMET assessment

In this study the Lupeol and the control molecules (MRP, and IMP) were retrieved from the PubChem database. The obtained natural compounds were filtered using SwissADME based on the rule of five through an online tool available at https://www.swissadme.ch/⁹¹. Also, we examine the toxicities related to organs such as AMES, hepatotoxicity, cardiac muscle toxicities, and their binding with the CYP cytochromes present in the liver for metabolism of the molecules tested through SMILES format using online server pkCSM-pharmacokinetics available at https://biosig.lab.uq.edu.au/pkcsm/⁹².

In-vitro antibacterial activity

The antibacterial activity was performed with the help of agar-well diffusion, and minimum inhibitory concentration (MIC), and minimum bactericidal concentration (MBC) were calculated. Freshly prepared methanolic CGL extract, Lupeol, and two control drugs (IPM and MRP) were used in all antimicrobial assays. The antibacterial activities were evaluated in a triplicate manner to assess greater efficacy.

Preparation of standard solutions

The methanolic CGL extract was used in the antibacterial activity by the adjustment of a standard concentration concerning the natural drug. The natural drug, Lupeol (25 mg) was procured from Sigma Aldrich (L5632) and was dissolved in HPLC grade 99.99% v/v methanol to provide a standard stock solution of 0.5 mg/ml. The stock standard solution was diluted sequentially with methanol to prepare the working solution at a starting concentration of 0.4 mg/ml. The CGL extract and Lupeol were used at the same concentration for better results.

Determination of zone of inhibition (ZOI)

The freshly prepared nutrient broth was inoculated with the bacterial colony and incubated for 3 to 4 h in the shaker at 180 rpm/min. The agar well diffusion method was carried out according to Zahid et al.⁹³ with slight modification. A fresh microbial culture of 100 μL of each inoculum was spread on Mueller Hinton Agar (MHA) plates for antibacterial assay, preparing 6 mm wells. The wells were incubated with CGL extract, Lupeol, and standard antibiotics (IMP and MRP) as control. After overnight incubation at 37 °C, the antibacterial activity was determined by measuring the diameter of the inhibitory zones and evaluated in a triplicate manner to assess greater efficacy.

Assessment of Minimum inhibitory concentration (MIC) and Minimum bactericidal concentration (MBC) of Lupeol, CGL, IMP, and MRP

The MIC and MBC of control antibiotics, CGL extract, and Lupeol solution were determined utilizing the 96-well microtiter plates and the micro broth dilution procedure. Additionally, the MBC value was calculated by subculturing the bacterial culture from each well of the microplate on nutrient agar plates at a predetermined amount of dilution⁹⁴. No microbial growth on the specific growth media was identified. The antibacterial activities were evaluated in a triplicate manner to assess greater efficacy.

Determination of cell wall disruption using Field Emission- scanning electron microscopy

The antibacterial activity was better understood by performing a microscopic study of the cell wall rupture experiment and visualizing the results on a field emission-scanning electron microscope (FE-SEM). Both treated and untreated samples were incubated for the entire night after bacterial pellets were treated with Lupeol solution. Following incubation, the pellets underwent three rounds of washing with 1 M phosphate buffer solution (PBS, pH 7.0), fixed with 2.5% glutaraldehyde, and then incubated overnight. Following centrifugation, the samples were cleaned using sodium phosphate buffer. Various ethanol concentrations of 30%, 50%, 70%, 80%, 90%, and ultimately 100% were used for washing the pellet and were subsequently dehydrated. After an hour of incubation at room temperature, pellets were dried and sent for SEM analysis⁹⁵.

In vitro enzyme inhibition assay

An in vitro enzymatic activity was performed to determine the effect of Lupeol on NDM-1 and to evaluate the 50% NDM-1 enzyme activity. The assays were carried out in 96-well plates containing 50 mM HEPES buffer, NDM-1 protein (0.016 mg/ml), and substrate nitrocefin 25 µM. Taking into the consideration of the MIC ranges of EDTA and Lupeol, EDTA (0.04 mg/mL to 0.0001 mg/ml) and Lupeol (0.04 mg/mL to 0.0001 mg/ml) activities were observed where, EDTA was used as a positive control. In brief, the NDM-1 enzymatic solution was incubated with different concentrations of Lupeol (0.04 mg/mL to 0.0001 mg/ml) in 50 mM HEPES (pH 7.5) at 37 °C for 15 min. The reaction was initiated by adding substrate nitrocefin, and NDM-1 activity was recorded by using the absorbance at OD 492 nm ⁹⁶. The IC₅₀ values for each inhibitory effect were analysed and interpreted using GraphPad Prism 8.0. The antibacterial activities were evaluated in a triplicate manner.

Evaluation of zinc supplementation’s efficacy

Zn²⁺ plays an important role in the hydrolytic activity of NDM-1. To evaluate the sensitivity of the compound to NDM-1 in the presence of Zn²⁺, we performed a Zn²⁺ inhibition assay⁹⁷. The experimental group includes Lupeol (0.02 mg/mL) as the test sample, and EDTA (0.03 mg/mL) served as the positive control. The assay was performed as follows, above the NDM-1 enzyme activity protocol, with the reaction initiated in the presence or absence of an equimolar concentration of ZnCl₂ (0.03 mg/mL). After incubation at 37 °C for 1 h, changes in absorbance at OD 300 nm were monitored for inhibition activity. Results are presented as the mean ± SD of triplicate samples.

Anticancer evaluation

Cell lines and culture

The Lungs cancer cell line (A549), ovarian cancer cell lines (OVCAR-3), Breast cancer cell line (MCF-7), prostate cancer cell line PC3 and Human embryonic kidney cell line (HEK-293) were obtained from Sigma Aldrich and ATCC. Cells were cultured and maintained in DMEM and RPMI 1640 media supplemented with 10% fetal bovine serum (FBS), 1% penicillin-streptomycin, and 2 mM L-glutamine. Cultures were incubated at 37 °C in a humidified atmosphere with 5% CO₂.

Cytotoxicity assay

To assess cytotoxicity, cancer cells and normal control cells were treated with varying concentrations of Lupeol (0.012– 0.4 mg/ml) for 48 h. Cell viability was evaluated using the Sulforhodamine B (SRB) assay as previously described⁹⁸. Results are presented as the mean ± SD of triplicate samples.

Nuclear morphology analysis

To evaluate nuclear morphology alterations induced by the Lupeol, A549 cells were seeded onto Cellvis glass bottom dishes (35 mm) and incubated for 24 h. After treatment with the Lupeol and standard drug Adriamycin at the IC50 concentration (determined in the cytotoxicity assay), cells were incubated for 48 h, fixed with 2% formaldehyde, and stained with 10 µM Acridine Orange and Ethidium bromide for 15 min. After washing with 1X PBS, fluorescence images were captured using a Nikon Eclipse Ts2R-FL microscope and analyzed using ImageJ software⁷².

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CGL

C. grandis leaves

GC-MS

Gas Chromatography-Mass Spectrometry

MBL

Metallo-β-lactamase

PDB

Protein Data Bank

IPM

Imipenem

MRP

Meropenem

VMD

Visual Molecular Dynamics

PCA

Principal Component Analysis

MHA

Mueller Hinton Agar

RMSD

Root Mean Square Deviations

RoG

Radius of Gyration

MIC

Minimum Inhibitory Concentration

MBC

Minimum Bactericidal Concentration

ZOI

Zone of Inhibition

Author contributions

S.L : Study design, data retrieval, *In-vitro* experiments, writing original draft, investigation. S.A.M: *In-silico* investigations, data curation, validation, writing, review and supervision. R.K.M: Cytotoxicity analysis, data curation, validation. B.S.D: In-vitro analysis, data curation and validation. S.K.S: Supervision, review, formal analysis. B.N: Formal analysis, investigation and review. S.K.T.: Formal analysis, investigation and review. D.D: Conceptualization, study design, supervision. All authors have read and agreed to the published version of the manuscript.

Funding

Open access funding provided by Siksha 'O' Anusandhan (Deemed To Be University). This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

We declare that all the data generated are included in this study and will be available on request to the corresponding author and the nucleotide sequences generated during this study was submitted to GenBank can be accessed through: https://www.ncbi.nlm.nih.gov/genbank/ with accession number for the *Klebsiella pneumoniae* : OP889580 are available at NCBI database.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

This article does not contain direct studies with human participants or animals performed by any of the authors. It contains microbial sources of swab samples collected from humans and the institutional ethical committee has approved the study (Ref. no/IEC/IMS.SH/SOA/2022/403). The required consent to participate was taken.