Abstract
Glioblastoma is the most prevalent type of brain tumor, and because of drug resistance, treatment for gliomas has been less successful. Citronellol is an acyclic monoterpene alcohol with various pharmacological properties. This study aimed to evaluate the effect of citronellol and its nanoformulation on glioblastoma cell proliferation. The physicochemical properties of citronellol and its synthesized silver nanoconjugates (CN@AgNPs) were evaluated using DFT and ADMET studies. The targets of the investigation (p53 and CDK4) were identified through the application of chemogenomics and analysis of the STRING protein-protein interaction network. Ligands were docked to the interaction sites of specific targets using AutoDock Vina 1.5.7. Molecular dynamics were used to mimic the citronellol complex CDK4 and p53. Because metallic bonds, which provide metals with unique strength and stability, are more resilient and long-lasting than hydrogen bonds, the results showed that the CN@AgNPs generated a more stable complex. Citronellol and CN@AgNPs were assessed by an in vitro study to determine the expression of IC50 concentration for the top scored selected genes to confirm the cytotoxicity of the compound against the GBM cell line SF-767. The findings showed that Citronellol and CN@AgNPs had concentration-dependent cytotoxic effects. Citronellol and CN@AgNPs, with IC50 values of 20.04 ± 4 µg/mL and 19.67 ± 4 µg/mL, respectively, decreased CDK4 expression and raised p53 expression in the SF-767 cancer cell line. In conclusion, the cytotoxicity and inhibition index of glioblastoma cells were increased by the phytocompounds coupled with AgNPs. Therefore, CN@AgNPs may be a good choice for treating cancer.
Introduction
About one-third of all brain malignancies are glioblastoma multiforme (GBM), the most prevalent and deadly type of primary intracranial tumor1. The median patient survival time is typically 10–12 months and the prognosis is poor2,3. Glioblastoma multiforme (GBM) is a serious and incurable brain tumor. Currently, chemotherapy with temozolomide (TMZ) remains the mainstay of glioblastoma treatment, but it has several drawbacks, such as undesired toxicity, drug resistance, and the blood-brain barrier (BBB)4,5,6. These problems underscore the pressing need for innovative multi-targeted treatment strategies7. The management of GBM is made more difficult by intratumoral heterogeneity, cancer stemness, cell-cycle dysregulation, and epigenetic changes. Cell cycle dysregulation and epigenetic changes have become actionable targets, providing optimism for more potent therapeutic approaches, although heterogeneity and stemness are currently unavoidable8.
Cell cycle dysregulation is an actionable pathway in glioblastoma that is crucial for tumor development. Two primary proteins, cyclins and cyclin-dependent kinases (CDKs), regulate cell cycle progression. The cell cycle stops, and mutagenic lesions are unable to spread to daughter cells when DNA damage occurs during the G2/M phase. p53 downregulation of p21 during DNA damage monitors these transitions9,10. Natural substances, such as phytochemicals, have drawn attention for their capacity to alter the p53/p21 pathway and other important targets, because of their crucial role in glioblastoma.
Novel pharmacologically active chemicals used in medicinal treatments are mostly derived from natural products11. According to reports, compounds obtained from medicinal plants can help treat a variety of illnesses, including pain, HIV/AIDS, Alzheimer’s disease, cancer, and malaria12. Plant essential oils (EOs), also known as volatile oils, are secondary metabolites produced by aromatic plants and are widely used in the agriculture, biomedicine, cosmetics, pharmaceuticals, and food industries. Recent studies have shown that EOs possess a broad range of biological activities, including antibacterial, antioxidant, and anticancer properties13. Essential oils have emerged as promising therapeutic agents against glioblastoma (GBM), exhibiting significant DNA-binding affinity and cytotoxic effects in cancer cell lines. These findings highlight their potential as natural anticancer compounds, and suggest their utility in both direct tumor targeting and as components of novel drug delivery strategies for GBM treatment14.
Citronellol, an essential oil, is a monoterpene with a molecular formula of C10H20O15. It serves as a guide for chemoprevention and as a chemotherapeutic agent for cancer treatment. In patients receiving chemotherapy and/or radiation therapy, citronellol has been shown to enhance immune function by decreasing leukocyte and neutrophil depletion and minimizing adverse symptoms, including nausea, dysgeusia, numbness in the limbs, and hearing loss16. Additionally, citronellol has been shown to reduce the effectiveness of medication therapies by blocking the activity of multidrug-resistant P-glycoproteins17.
However, it is typically unknown how phytocompounds are bioavailable and processed metabolically, and the blood-brain barrier frequently limits the effectiveness of glioma therapy. The bioavailability and targeted delivery of therapeutic drugs within the bulk of GBM tumors can now be enhanced by loading them onto nanoparticles (NPs)18. In vitro, these platforms demonstrate enhanced cellular absorption and depth of penetration into tissues19. Nanoparticle-based formulations provide a stable, reasonably priced, and biocompatible technology that is simple to scale and improves the solubility of medicinal compounds that are not very soluble20.
The difficulty in precisely focusing on the p53/p21 pathway because of factors such as TP53 mutations and CDK4 over-activation highlights the necessity for creative methods to control their activity. In this study, we created citronellol-based nanoconjugates and used a combination of computational and experimental methods to assess their ability to affect TP53 and CDK4. This study aimed to determine the molecular mechanisms behind the therapeutic advantages of these nanoparticles and assess their potential as innovative therapies for glioblastoma.
Materials and methods
Materials
Carboxy methylcellulose (CMC), Citronellol with 95% purity, Silver nitrate (AgNO3) and Sodium hydroxide (NaOH) were acquired from Sigma-Aldrich (Merck Group, Germany). All the reagents were analytically pure and did not require additional purification.
Ligand retrieval and its nanoformulation
An examination of the literature led to the selection of citronellol for the design of a citronellol-silver nanoconjugate (CN@AgNPs). With a few changes, the citronellol-silver nanoconjugate was synthesized according to Ali et al.‘s20 description. Deionized water (300 ml) and CMC (7 mg) were added to a 500 mL flask. Citronellol (3 ml) was added and the mixture was mechanically stirred for 15 min while spinning at 70 °C to produce a uniform viscous solution. 200 ml of a 0.05 M solution was prepared by adding AgNO3 to deionized water and stirring for five minutes. The initial solution was supplemented with this one. After gradually adding NaOH to obtain a pH closer to that of the base, the liquid was stirred for 15 min at 100 rpm using a magnetic stirrer. The mixture turned brown instead of colorless, indicating the production of CN@Ag NPs. A portion of the solution was centrifuged for 20 min at 5000 rpm and then allowed to dry for 48 h at room temperature in order to facilitate additional characterization. In vitro analysis was performed using the remaining quantities.
Characterization of Citronellol-based silver nanoconjuagte (CN@AgNps)
Particle size (PS) and zeta potential (ZP) measurements
The electrostatic potential and average PS of the CN@AgNPs have been described and are useful parameters for stability studies21. An Anton Paar Zetasizer Nano ZS (Litesizer 500) was used to measure the PS and ZP of the conjugated AgNPs produced at 25 °C following dilution with dH2O. To reduce the chance of errors, the trials were independently performed three times22,23.
Ultraviolet-Visible absorption spectroscopy
A Shimadzu UV-Vis spectrometer (UV-1800, Shimadzu, Japan) was used to evaluate the absorbance of the samples. This confirms the quality of the content. Spectral analysis verified that the CN@AgNps were produced. The visible and ultraviolet spectral bands (200–800 nm) were used to log the CN@AgNPs spectra. Distilled water was used as a reference for baseline adjustment24.
Fourier transform infrared spectroscopy (FT-IR) analysis
To determine chemical bonding, IR spectra were obtained using a Shimadzu FTIR spectrometer. The functional groups that were in charge of the stability and reduction of the biosynthesized AgNPs were identified using FTIR analysis. In this study, wavenumbers ranging from 400 to 4000 cm − 1 were employed25.
X-ray diffraction (XRD) measurement
Using a Shimadzu XRD-6000 diffractometer (Japan), the experimental powder’s purity and crystal structure were evaluated. All scans were performed at room temperature using Cu Kα radiation, which corresponds to λ = 1.056 Å at diffraction angles of 10–70 °26.
Scanning electron microscopy (SEM)
An FEI Nova NanoSEM 450 (USA) microscope was used for morphological examination. SEM was used to confirm the size, shape, and crystallinity of CN@AgNPs27.
In Silico Analysis
Density functional theory (DFT) studies
DFT calculations were performed using a slightly modified version of a previously reported method 28. All computations in the split-valence polarization (SVP) basis set were performed with the default parameters using the B3LYP function in the Gaussian 06 package (Rev. E.01). Efficient computation of atoms’ and molecules’ electronic structures is made possible by this theory. The best geometric parameters, global and local reactivity descriptors, frontier molecular orbitals (FMO), and molecular electrostatic potentials (MEP) were identified. The checks were conducted using the Guass View 629.
Pharmacokinetic parameters
Furthermore, the absorption, distribution, metabolism, and toxicity (ADMET) characteristics of the newly created CN@AgNPs and citronellol were predicted using SwissADME (http://www.swissadme.ch)30. ADMETlab 2.0 (https://admetmesh.scbdd.com/)31 was used to estimate the pharmacokinetic properties of the compounds. These investigations evaluated attributes such as the extent of absorption, distribution volume, CYP-binding metabolism, excretion, and AMES toxicity prediction.
StopTox (https://stoptox.mml.unc.edu/)32 was used for the toxicity studies. After gathering, choosing, and integrating the largest publicly available datasets, we assessed the toxicity of the compounds for various toxicity endpoints.
Compound target predictions for glioblastoma via PIDGIN protocols
Following pharmacokinetic prediction, the target of citronellol was predicted using chemogenomics software Pidgin version 2 with the Python command “predict_per_comp.py input.csv 30 0.5 0.3 “Homo sapiens (Human)“33,34. The Python script was altered to produce binary forecasts, where a “0” or “1” denotes activity or inaction, respectively. Using input vectors of length 2,048 bits and radius 2, hashed circular Morgan fingerprints were created using the RDKit35. These fingerprints, known as ECFP_4 (an Extended Connectivity Fingerprint with a default maximum diameter of four for the circular neighbors of each atom), served as the basis for the models. The models generate probabilities that show the possibility of compound-target interaction after being trained on the binary properties of both active and inert compounds. These probability values are then converted into binary forecasts using target-specific thresholds36,37.
Validation
GeneCards (The Human Gene Database) https://www.genecards.org/ and MalaCards (Human Disease Database) https://www.malacards.org/ were used to confirm PIDGIN predicted gene targets for glioblastoma. The GeneCards database was used to annotate the molecular functions of probable targets and biological processes38. MalaCards is a database resource for human diseases, and the annotations associated with them are accessible online38. The GeneCards database was created using GeneCards architecture and methodology39.
Protein-Protein interaction (PPI) network construction and KEGG enrichment analyses
Gene function has often been annotated and predicted using protein-protein interaction (PPI) data, assuming that interacting proteins have comparable or equivalent activities and may thus be involved in the same pathways. Similar biological functions are usually carried out by nearby proteins in a PPI network40. To explore the molecular pathways and functional linkages of the targeted proteins identified by PIDGIN, the first weighted PPI network was constructed using STRING data41.
To identify pathways related to glioblastoma and comprehend the regulatory function of the targeted protein in processes such as cell cycle progression and its relationships with nearby molecules in the network, KEGG pathway analysis42,43 was performed using STRING44.
Protein preparation
Using BIOVIA Discovery Studio Visualizer 2021, the receptors were designed by removing native ligands, heteroatoms, and water molecules. Polar hydrogen atoms and Gesteiger partial charges were added to the pdbqt protein file using AutoDock Tools, version 1.5.745,46,47.
Active site identification
The “receptor cavity method” was employed to forecast the receptor proteins’ binding sites using BIOVIA Discovery Studio 2021. The BIOVIA Discovery Studio SDB-Site module enabled the identification and description of protein structural binding sites. In this process, the inhibitory properties of the residues with center-x, y, and z values that were present in the binding sites were investigated48.
Ligand Preparation
The structural data file (SDF) format of the NCBI PubChem database (https://pubchem.ncbi.nlm.nih.gov/) was used to obtain the citronellol and reference drug temozolomide structures. The 2D structures of CN@AgNPs were manually designed using ChemDraw Professional 16. Chem3D 16.0 was then used to convert the 2D structures into 3D structures49,50. The Chem3D Gaussian interface was used to minimize energy consumption and save the files as an SDF. The SDF files of both ligands underwent ligand processing to assign the proper bond ordering51.
Molecular Docking
AutoDock Vina 1.5.7 was used for molecular docking46. After preparation, the ligand and receptor structures were moved to the Vina folder and stored in the pdbqt format. Vina’s AutoDock application was launched using a Command Prompt (CMD)45. The programming command that was executed was “vina.exe --config conf.txt --log log.txt”45. This approach computes binding affinity using a grid-based model of protein-ligand potential interactions. Vina Tools’ soft-core potentials have been shown to be useful in producing a variety of random conformations of small organics and macromolecules within the target protein’s active region. To identify potential treatment alternatives, ligands were docked to the proteins, and their relative degree of interaction was evaluated47. The ligand interaction tool Discovery Studio 2021 was used to view the interaction diagram of the active-site residues of ligand proteins52,53.
Molecular dynamic simulation
Molecular dynamics (MD) simulations were conducted on relevant ligand-protein complexes to examine the stability of ligand binding54. To assess the dynamic qualities of the complex, it was loaded into the system and exposed to ff14SB in amber. The ligand was simultaneously subjected to a generalized AMBER force field. Proteins that contained protonation were neutralized using the LEaP method and counterion 2Cl-. At an edge distance of 9.0, the system solved this problem. Coordinates and parameters were generated using the LEaP approach, and the resulting compound was saved in PDB format. The amount of stearic acid was reduced by a factor of three times in order to eliminate its effects. During the first minimization step, solvation and ionization were used to optimize proteins and ligands. The optimized pocket residues included the backbone amino acids and proteins. During the last minimization phase, the complete system was activated to facilitate protein complexation. Following the minimization stage, the system was installed within the heating panel, where the temperature was gradually increased. MD production of 100 ns using the NPT ensemble at 1 atm and 300 K was performed after the system had stabilized. Finally, once the MD simulations with specific complexes were completed, certain metrics, such as RMSD and RMSF, were analyzed using another module, CPPTRAJ, of the Amber20 software55,56. We also calculated the radius of gyration (Rg) and dynamic cross-correlation matrix (DCCM) for 100 ns of poses using the Arantes et al. approach57. MMPBSA/MMGBSA modules were integrated with Amber17 to calculate binding free energy (BFE)58.
In vitro analysis
Material
Chemicals and reagents used in this study were purchased from Sigma-Aldrich (Merck, Germany).
Cell line
Human glioblastoma SF-767 cell lines was obtained from cultural lab of The University of Lahore, Lahore, Pakistan.
Cell culture
The cells were maintained in an incubator with 5% CO2 at 37 °C in Dulbecco’s modified Eagle’s medium. Next, 100 units/ml penicillin, 100 µg/ml streptomycin, 1 mM sodium pyruvate, and 1 mM nonessential amino acids were added along with 10% fetal bovine serum59.
Treatment
To achieve 80 ± 5% confluence, the cells were grown at a density of 104 cells/well in 96-well plates at 37 °C, 21% O2, and 5% CO2. Next, 50 µL of fresh medium was added to each well instead of the liquid medium. The cytotoxic effects of CN@AgNPs and free citronellol were investigated at dosages of 3.13, 6.25, 12.5, 25, 50, and 100 µg/mL by adding 100 µL/well of successive dilutions. For the control group, 100 µL of dimethyl sulfoxide (DMSO) was added to each well. For twenty-four hours, the treated plates were incubated60.
Cell viability and cell death
The MTT assay was used to evaluate cell viability following treatment. The plates were incubated for four hours after 50 µL/well of MTT solution (0.5 mg/m/L) was added (the chamber was filled with 60% oxygen at normal pressure). To dissolve the formazan crystals, 200 µL/well of DMSO was applied after the liquid media was disposed of. A plate reader was used to measure the optical density (OD) of the wells at 570 nm, and cell viability was calculated. The tests were performed thrice, and the means and standard deviations were determined60.
Effect of compounds on expression of P53 and CDK4 in Glioblastoma (SF-767) realtime- polymerase chain reaction (RT-PCR)
Following the manufacturer’s recommendations, RNA was extracted using TriPure reagent, and a UV spectrophotometer was used to determine the concentration. For RT-PCR, five micrograms of extracted RNA were utilized. Mouse leukemia virus reverse transcriptase was inactivated for five minutes at 95°C. After 40 cycles of denaturation at 95°C for 30 s, annealing at 59°C for 30 s, extension at 72°C for 30 s, and extension at 72°C for 10 min. As an internal control, β-actin was employed. The data were examined by 1.5% agarose gel electrophoresis, software processing, and ethidium bromide staining61. In this experiment primers used P53 (152 bp), 5’- TACTCCCCTGCCCTCAACAA-3’ (forward) and 5′- CAGCGCCTCACAACCTCC-3’ (reverse); CDK4 (205 bp), 5′- AGTGGCTGAAATTGGTGTCG-3′ (forward) and 5′- ACCTTCTCACCGGACAACAT-3′ (reverse).
Statistical analysis
GraphPad Prism software (GraphPad, San Diego, CA, USA) was used to statistically evaluate the data (mean ± SD) with a 95% confidence interval60,61. The non-linear regression parameter was used to determine the IC50 values60.
Results
Citronellol-based silver nanoconjuagte (CN@AgNPs) and their characterization
Figure 1 shows the procedure for the production of AgNPs based on citronellol. According to the chemistry underlying the creation of the nanoparticles, citronellol’s hydroxyl groups (-OH) attract Ag ions, starting the process of converting them into silver nanoparticles. CMC acts as a stabilizing agent, while citronellol controls the size of the nanoparticles by capping and reducing them. NaOH was gradually applied to reduce the pH to basic levels. The solution was neutralized by hydroxyl ions (OH −) from NaOH, which also promoted the reduction of Ag + to metallic Ag (Ag⁰) nanoparticles. When the hue changed from translucent to brown, citronellol-based AgNPs were formed.
Preparation Steps for CN@AgNPs (change of color from transparent to brown showed that nanoconjugation occured).
Particle size (PS) and zeta potential (ZP) measurements
The produced AgNPs had a mean size of 150.88 nm, according to particle size investigations, and with a mode of particle distribution of 169.08 nm with a standard deviation of 62.19 nm (Fig. 2A) (Supplementary Figure S1). Additionally, the results demonstrated that the polydispersity index (PDI) of 0.17 for the biosynthesized NPs was sufficient for drug transport applications62. Monodispersed stable nanoparticles are characterized by a PDI of less than 0.5, as this metric measures the stability and dispersity of a particle63. Conjugated AgNPs were found to have a decreased zeta potential (-1.7 mV), indicating that the produced nanoparticles were less stable and more susceptible to assembly (Fig. 2B) (Supplementary Figure S2).
Characterization of Citronellol-based Silver nanoconjugates: Particle size determination: Synthesized nanoparticles size distribution (A) and analysis of zeta potential of synthesized silver nanoconjugates ( B), Ultraviolet-visible spectroscopy: SPR bands spectrum of CN@AgNPs (C).
Ultraviolet-Visible spectrometry
UV-Vis spectroscopy was used to confirm the synthesis of the nanoparticles and to assess the correlation between the average particle diameter and the peak observed at a specific wavelength. Additionally, the widths of the Surface Plasmon Resonance (SPR) bands were used to confirm the dispersion of the nanoparticles (Fig. 2C). In Fig. 2C, the nanoparticles (NPs) show a prominent peak at approximately 305 nm, which corresponds to a single surface plasmon resonance (SPR) band typical of spherical nanoparticles. This is further supported by the band peak at a lower wavelength, indicating that the NPs are comparatively small in size (Supplementary Figure S3).
Identification of functional groups using fourier transform infrared spectroscopy (FT-IR)
Sharp absorption bands at 1347, 1642, and 3301 cm−1 were visible in the FTIR spectra of CN@AgNPs (Fig. 3). These correspond to functional groups also seen in pure citronellol, were almost similar with minor shifts, and indicate the role of citronellol as a conjugated agent to AgNPs.
FTIR spectrum: Graph with wavelength (cm−1) on the x-axis and transmittance percentage on the y-axis.
A broad peak observed around 3312 cm⁻¹ in both spectra corresponds to O–H stretching vibrations, with a noticeable broadening and slight shift in CN@AgNPs, suggesting hydrogen bonding or possible coordination between the hydroxyl oxygen and silver ions, indicating successful surface interaction. A significant shift in the C = O stretching band from 1636 cm⁻¹ in citronellol to a lower intensity and a slight shift in the CN@AgNPs further support the coordination of carbonyl groups with silver ions. Moreover, the C–H bending vibration peak near 1366 cm⁻¹ in citronellol is retained in the CN@AgNP spectrum with minor changes in intensity, indicating that the core structure of citronellol remains intact after conjugation. A distinct band at 2921 cm⁻¹ (C–H stretching) appeared in citronellol and was subtly reduced in intensity in CN@AgNPs, corresponding to an alkyl-saturated aliphatic group.
X-ray diffraction (XRD) measurement
The crystalline nature of the AgNPs was clarified by examining the XRD data. Citronellol-conjugated AgNPs are evident from the distinctive peaks observed in the XRD pattern (Fig. 4). At 2theta values of 39°, 46°, 66°, and 79°, distinct diffraction peaks correspond to the (111), (200), (220), and (311) planes, respectively. This suggests that AgNPs are crystalline andface-centered cubic (FCC) in nature (JCPDSfile no. 04-0783)64. Additionally, a few peaks (represented by stars) were observed, suggesting that citronellol interacted with the surface of the silver nanoparticles.
XRD graph of CN@AgNPs, showing Peak Indices & 2 θ Positions.
Scanning electron microscopy (SEM)
SEM examination was performed on the samples to determine their particle size distribution and shape. According to the SEM analysis, the AgNPs made from citronellol were spherical in shape. A histogram was constructed to determine the number of AgNPs in the chosen area (Fig. 5A). An average particle diameter of 175 ± 0.015 nm was obtained using nonlinear curve fitting (Gaussian model). The compositions of the samples were investigated using SEM and energy-dispersive X-ray spectroscopy (EDX). Each element had a corresponding number of X-ray counts according to the net intensity values derived from Energy Dispersive X-ray (EDX) analysis. Higher percentages of pure AgNPs were detected using EDX (Fig. 5B). Compared with the other peaks, the silver peak was higher. EDX analysis demonstrated that the sample contained the necessary phase of silver (Ag). This is most likely because the NP sample was held over the substrate during SEM imaging. The higher weight% of silver and the higher net intensity of silver (4736.31) relative to oxygen (385.63) indicate that silver was more common in the sample. The longer peaks of the oxygen peaks from the EDX of the nanoconjugate may further suggest the existence of the OH group, as they show that citronellol was adsorbed on the nanoparticles (Fig. 5B).
SEM image of CN@AgNPs: Magnified image of selected area with histogram (A) and EDX spectrum (B).
In Silico analysis
Density functional theory (DFT) studies
Frontier molecular orbital analysis (HOMO-LUMO) was used to predict the reactivity of citronellol and its synthetic Ag conjugates (CN@AgNPs) in comparison with the reference drug temozolomide (Fig. 6). The frontier molecular orbitals (FMOs) of a molecule, such as HOMO and LUMO, dictate how they interact with other molecules. The energy gap (∆Egap) also helps describe the chemical reactivity and kinetic molecular stability.
The anticipated dipole moment, lowest unoccupied molecular orbital (LUMO), highest occupied molecular orbital (HOMO), and total energies of citronellol and its silver nanoconjugates, along with the reference drug, are displayed in Table 1. Since the total energy of citronellol-silver nanoconjugates is lower than that of free citronellol and temozolomide, AgNPs are more stable than free ligands. The HOMO energy indicates a compound’s ability to donate electrons, while the LUMO energy describes a compound’s ability to receive electrons. The kinetic stability and chemical reactivity of a molecule can be explained by its HOMO level. The energy gaps (Eg) = ELUMO − EHOMO were lower for CN@AgNPs than for free citronellol because of the ligand-to-metal-ion chelation (Table 1; Fig. 6). The lower Eg values of the complexes compared with those of the free ligands provide an explanation for the charge-transfer interactions that occur during complex formation. Electrophilicity (ω), electrical potential (µ), softness (S), hardness (η), electronegativity (χ), ionization energy (I), and electron affinity (A) were computed (Table 1).
Frontier molecular orbitals (FMOs) of citronellol (A), CN@AgNPs (B) and Temozolomide (C) in the form of HOMO, LUMO and the Energy gap (Eg) using the Gaussian 06 package.
Molecular electrostatic potential (MEP) studies
MEP surface designs in which the positive and negative sectors are electrophilic and nucleophilic areas, respectively, have also been used to characterize the reactive behavior of a molecule. In this study, MEP maps (Fig. 7) revealed regions of positive (blue) and negative (red) electrostatic potentials on the molecular surfaces. This indicates potential sites for electrophilic and nucleophilic interactions. The targets under investigation showed a wide distribution of both positive and negative potential regions across their molecular skeletons, suggesting that they might be closely linked to their respective molecular electrostatic potentials. This result may help in designing a stabilizing complex for the best drug docking within the ligand binding of the targets being studied.
Molecular electrostatic potential surface of citronellol (A), CN@AgNPs (B) and Temozolomide (C) using the Gaussian 06 package.
Pharmacokinetic parameters
Analysis of physicochemical properties and drug-likeness
The physicochemical and drug-likeness properties of citronellol, CN@AgNPs and reference drug temozolomide were investigated using SwissADME. The physicochemical properties are presented in Table 2. Citronellol and CN@AgNPs had respective TPSA (total polar surface area) values of 20.23 Å and 9.23 Å, suggesting that the phytochemical and its produced nanoconjugates were within the cut-off range of the Lipinski rule of five (LRF). Among the tested compounds, CN@AgNPs showed the lowest polar surface area (9.23 Ų), indicating superior membrane permeability followed by CN (20.23 Ų), while the reference drug (108 Ų) exhibited moderate permeability. The drug resemblance of both candidates was further demonstrated by the fact that none of the other metrics used by Ghose et al. had any violations (apart from temozolomide and citronellol’s Ghoose violation).
ADME analysis
The radar plots in Fig. 8 (A, B and C) show that all the related parameters, that is, flexibility, lipophilicity, size, polarity, insolubility, and instauration, fell within the specified red zone (except temozolomide’ insaturation property), indicating a good candidate for oral drug delivery. The primary ADME descriptors involved in the absorption, distribution, metabolism, and excretion of drugs are displayed in Table 3.
The boiled egg and the absorption of both ligands by the body are depicted in Fig. 8 (D). Based on the BOILED-Egg model, CN@AgNPs and citronellol fall within the blood–brain barrier (BBB) permeable region, indicating strong CNS penetration potential, unlike the reference drug. As the figure shows, these substances are not P-glycoprotein (PGP) substrates, and they were not associated with an increase in drug resistance. Additionally, the phytochemical citronellol and its nanoconjugate can cross the blood-brain barrier (BBB) because of their small size.
Bioavailability radar of selected compounds citronellol (A) and CN@AgNPs (B) and Temozolomide (C) using the SwissADME server, BOILED-Egg results of selected compounds in comparison, generated by SwissADME server (D) and the association between the selected chemicals and endpoint toxicity is shown in a bar chart (E).
Toxicology profile
It is necessary to take toxicological effects into account when assessing the possible harm that an inhibitor could cause to the human body. The ADMETlab server was used to perform the toxicophore tests. All the selected drugs exhibited higher potential as therapeutic candidates with zero alertness, as indicated by the toxicological profile data in Table 3. Both ligands exhibited notable non-toxic behavior against all evaluated parameters according to the Stoptox toxicity investigation (Fig. 8E), whereas the reference drug, temozolomide, showed significant oral acute toxicity (71%) and eye irritation/corrosion potential (82%).
Compound target predictions for glioblastoma via PIDGIN protocols and validation
PIDGIN v2 (Prediction Including Inactivity), a program that uses ECFP4 circular Morgan fingerprints trained on ChEMBL actives and PubChem inactives, was used for in silico target prediction to identify possible active targets for citronellol. Eight of the 343 protein targets identified in our chemical analysis (Supplementary File S1) were first linked to glioblastoma: IDH1, TP53, BCHE, MMP2, PTGS1, NFKBIA, RUNX1, and FAK.
The Human Gene Database (GeneCards) and Human Disease Database (MalaCards) were used to validate the predicted targets. Of the eight targets, MMP2 was also found in MalaCards, and IDH1 and TP53 were linked to glioblastoma in both databases. TP53 was selected for additional research based on the validation results because it had the best relevance scores in both databases (Fig. 9).
Validation of genes for glioblastoma; interpretation via GeneCards and MalaCards.
Protein-Protein interaction (PPI) network construction and KEGG enrichment analyses
The activation of genes linked to cell cycle arrest and apoptosis is facilitated by the p53 protein pathway. We constructed a PPI network of p53 in the STRING database to explore genes that are directly associated with p53. Five interacting proteins, CDK4, PTEN, CDKN2A, MDM2 and EGFR, were identified by TP53 analysis (Fig. 10A). The KEGG glioblastoma pathway (Supplementary Figure S4) indicated that CDK4 has a direct role in cell cycle progression, which is controlled by TP53 through p21 (CDKN1A). Glioblastoma is characterized by dysregulation of CDK4, which makes it an important co-target, in addition to TP53.
Gene Expression Profiling Interactive Analysis (GEPIA) was used to examine CDK4 expression in GBM (Fig. 10B). The fact that CDK4 expression is higher than that of TP53 suggests that it may be used as a target for drugs and diagnostics. In contrast to TP53, this gene expression study showed that CDK4 is a key marker of unchecked cell division during malignant development, supporting these findings.
(A) STRING Database PIP network of p53, showing protein-protein association; associations are meant to be specific and meaningful, i.e. proteins jointly contribute to a shared function. (B) The expression of these genes in GBM and normal controls was compared using Student’s t-test and presented as scatter-box plots using Gene Expression Profiling Interactive Analysis (GEPIA).
Molecular Docking
The action of citronellol and its nanoconjugate CN@AgNPs against the tumor suppressor p53 in glioblastoma was verified by molecular docking. The reference drug temozolomide was also included for comparison with the selected ligands. The binding affinities of all selected compounds were also evaluated for target CDK4 using molecular docking. Their potential as a dual-target treatment approach for glioblastomas was assessed. To assess the capacity of the hit to attach to targets, we created three-dimensional conformations for every molecule and docked them into the active site of the protein structures, p53 (PDB ID: 8DC8) and CDK4 (AF-P11802-F1-v4).
p53 protein structure emphasizes important functional domains such as the dimerization site, zinc-binding site, and DNA-binding domain. As a possible drug-binding site, the hydrophobic pocket created by the Y220C mutation is far from the location where p53 binds to DNA. Labeling structural elements such as coils, β-helices, and β-sheets makes it easy to see how the protein is organized and where the docking target is located (Fig. 11A). Graphical analysis revealed that residues L145, W146, V147, C220, P222, P223, and P151 were frequently found in the active site of the p53-ligand complex-bound system. Table 4 provides a summary of the binding affinities and modalities of interactions of the compounds with the target.
Structure of p53 protein showing the four domains (A), 3D image of selected ligands (B, C, D), 3D interaction diagram of p53 complex with citronellol (E), CN@AgNPs (F) and temozolomide (G).
The binding scores of citronellol (-4.6 kcal/mol) and CN@AgNPs (-4.9 kcal/mol) were comparable to that of the reference drug temozolomide (-5.5 kcal/mol). Similar to temozolomide (Fig. 11D, G), citronellol engages the hydrophobic pocket of the p53 binding site primarily through hydrophobic interactions with key residues. These interactions allow for stable binding without the formation of hydrogen bonds, indicating that the hydrophobicity of the binding pocket is essential for p53 restoration and ligand stabilization. (Fig. 11B, E).
However, the CN@AgNPs created a more stable complex by forging strong coordination connections with the oxygen atoms of the cysteine residue. The electron pairs that interacted with Ag were supplied by the O atoms in cysteine (Fig. 11C, F). CN@AgNPs are more stable toward protein p53 with metal-acceptor links with residues C220 because metallic bonds are stronger and more resilient than other bonds, which gives metals their unique strength and stability.
Structure of CDK4 protein showing the ATP-binding domains (A), 3D image of selected ligands (B, C, D), 3D interaction diagram of CDK4 complex with citronellol (E), CN@AgNPs (F) and temozolomide (G).
The structure of the CDK4 protein, which has an ATP-binding domain, was obtained from the alphafold (ID: AF-P11802-F1-v4) (Fig. 12A). Drug designs targeting the same pocket have been reported by several research groups. The CDK4 active site frequently contained residues I12, G13, E144, N145, L147, G15, A157, D158, G18, V20, A33, K35, V72, F93, E94, H95, V96, and D99. Table 5 summarizes the binding affinities and modalities of interaction between citronellol and its silver-based nanoconjugates with CDK4, in comparison with temozolomide.
When citronellol binds to the nearby CDK4 residues D158 and F93, two hydrogen bonds are formed. The hydroxyl group of aspartic acid (D158) (donor motif) and carbonyl oxygen (acceptor motif) of citronellol were found to interact through hydrogen bonds in the citronellol-CDK4 complex. Additionally, the hydroxyl group of citronellol (donor motif) and the π-orbital system of phenyl (F93) formed a hydrogen bond in the CDK4 binding pocket (Fig. 12B, E).
Furthermore, the CDK4 complex containing CN@AgNPs outperformed free citronellol in terms of its stability. This occurs because of the metal-acceptor bonds of the silver atoms in CN@AgNPs made with V96 (Fig. 12C, F). Furthermore, the stability of both complexes was enhanced by alkyl and pi-alkyl interactions (Table 5).
In contrast, the reference drug temozolomide formed a less stable complex with CDK4, as it was surrounded by fewer interacting residues, despite having a comparable binding affinity (-6.0 kcal/mol) (Fig. 12D, G). This suggests that citronellol and its nanoconjugates may offer improved binding stability over temozolomide.
Molecular dynamic simulation
The dynamic stability and intermolecular interactions of citronellol in complexes with p53 and CDK4 target proteins were evaluated as a function of 100 ns, using classical MD simulations. Figure 13 shows the root mean square deviation (RMSD) of citronellol in conjunction with CDK4 (AF-P11802-F1-v4) and p53 (PDB ID: 8DC8). RMSD values of the CDK4/citronellol inhibitor combination were evaluated during a simulated interval of 100 ns. The ligands, pockets, and protein RMSD were analyzed for dynamic stability and sampling patterns. RMSD was computed over 100 ns using the initial shapes of the molecules. The protein atoms, pockets, and ligands in the CDK4/citronellol complex remained constant (Fig. 13A). The ligand demonstrated even more stability, with an RMSD consistently below 2.5 Å, indicating that it was securely bound within the binding pocket with minimal mobility. Protein 8DC8 in Fig. 13B has an RMSD fluctuation of approximately 2 Å to approximately 3.5 Å, with an initial RMSD of 1.5 Å to 2.2 Å that gradually rose at 40 ns. With an RMSD between 0.5 and 1.5 Å, the ligand is stable inside the binding pocket.
The overlapped conformations of CDK4 showed few structural variations, with citronellol regularly interacting through hydrogen bonding and hydrophobic interactions with important residues, such as K35, V72, F93, and D158 (Fig. 13C). Significant structural changes are shown by the overlapping conformations of p53, where citronellol binds close to important residues, such as T60, T140, C130, and P132 (Fig. 13D).
RMSD graphs for citronellol in complex with p53 (B and D) and CDK64 (A and C); the ligand RMSD values were determined as the protein-fit ligand based on the unique 100 ns MD simulation trajectory of each docked complex. The protein RMSD values of the alpha carbon atoms were obtained.
A comparison was made between two protein-ligand complexes subjected to 100 ns molecular dynamics (MD) simulations. The analysis focused on three aspects: (A and B) RMSF values for each protein structure residue, (C and D) protein radius of gyration over time, and (E and F) protein SASA throughout the 100 ns period. Specifically, (A, C, and E) represent the data for the CDK4/citronellol complex, while (B, D, and F) correspond to the p53/citronellol complex.
Additionally, the root-mean-square fluctuation (RMSF) of the ligand–protein complexes was investigated. The RMSF values of citronellol in association with CDK4 and p53 are shown in Fig. 14(A, B). With the exception of the binding site, the majority of the RMSF values were less than 3 Å, suggesting that both proteins remained stable during the 100 ns simulation. While the N-terminus of the protein displayed fluctuations in the p53 complex, the C-termini of the protein tail fluctuated in the CDK4 complex. Nonetheless, a discernible variation was noted in CDK around residues 30–60 and 90–120 at approximately 4 Å. The fact that the p53 changes near residues 124–140 lay approximately 5 Å is noteworthy at the same time.
The protein-folding state, modification, and overall compactness were ascertained using the radius of gyration (RGyr). As shown in Fig. 14C, CDK4 receptor is compact and stable. There was a slight variation between 0 and 40 ns, from 20.2 Å to 20.8 Å. The Rg values for p53 varied from 16.4 to 16.9 Å (Fig. 14D) in contrast to CDK4, suggesting a slightly less compact structure.
The rather steady trend in SASA for CDK4 following the first phase (~ 20 ns) indicates that the protein’s exposure to the solvent changed very little, suggesting structural stability (Fig. 14E). As the SASA of the protein decreases after binding to a ligand, the lower SASA range indicates that citronellol causes compactness or reduces solvent exposure in p53 when compared to its unbound state (Fig. 14F). This suggests that a portion of the protein’s surface is buried in the binding interaction, which reduces exposure to the solvent.
The mobility of residues along the protein chain was correlated using a Dynamic Cross-Correlation Matrix (DCCM). The plots in Fig. 15 (A, B) show distinct DCCM patterns for both the complexes. A color-coded scheme was created to represent the degree of correlation between the nobilities; a low correlation with the residues is indicated by the blue color, whereas highly related mobility is shown by the red to pale green hues. The associated motions of complex A showed a balance between areas of high positive (red) and negative (blue) correlations, which could indicate flexibility and cooperative movements between certain residue pairs. The diagonal line (red) indicates self-correlation as the residues have a perfect connection with one another. A more unified and coordinated action throughout the protein structure seems to be displayed by Complex B. The lower negative correlation (blue) compared to that of Complex A indicates less dynamic flexibility.
Cross-correlation matrix showing coordinate fluctuations for Cα atoms around the mean positions during MD simulation: positive correlations are represented by red, whereas negative correlations are represented by blue. CDK4 (A) and p53 (B).
(A-F). Two main eigenvectors of citronellol with complex CDK4 (A-D) and p53 (E-H) Gibbs free energy landscape (FEL). Lower-energy conformations are shown by the purple region, meta-stable states are shown by the green region, and high-energy conformations are shown by the yellow.
Unlike PC1 and PC2, we examined the subconformational structural alterations in reteplase using the Gibbs free energy landscape (FEL), which illustrates the interaction between a protein’s free energy and the molecule’s three-dimensional structure. The 2D and 3D images of the FEL analysis for citronellol in CDK4 and p53 complexes, along with ΔG values ranging from 0.5 to 3 kJ.moL-1, are shown in Fig. 16 (A-H).
Binding free energy calculations
Using MM/PBSA analysis, Fig. 17 shows the binding free energy decomposition for the two systems, most likely CDK4 (A) and P53 (B). The main causes of binding in system A are Van der Waals forces (-26.13 kcal/mol) and electrostatic interactions (-11.06 kcal/mol); polar solvation adds a penalty (+ 18.98 kcal/mol), making the overall binding free energy − 18.20 kcal/mol (PB model). System B has a binding free energy of -15.21 kcal/mol (PB model), with solvation penalties being less noticeable and Van der Waals forces (-25.53 kcal/mol) still predominating but electrostatic contributions being smaller (-3.64 kcal/mol). Because of its advantageous hydrophobic and electrostatic interactions, System A generally showed a higher binding affinity than System B. These results were consistent with the docking results.
Comparison for the binding free energy terms applied, CDk4/citronellol (A) and p53/citronellol (B).
In vitro analysis
Cell viability and cell death
The cytotoxic activity of citronellol and the newly synthesized citronellol-based silver nanoconjugates (CN@AgNPs) on the human glioblastoma cell line (SF-767) was evaluated using an in vitro MTT assay. After 48 h of exposure to doses ranging from 3.13 to 100 µg/mL, the percentage of cells that died was computed in relation to the cells that remained unaffected (Fig. 18). The findings demonstrated that both compounds exhibited particular cytotoxicity against the cancer cell line SF-767, ranging from 7 to 32% when compared to the reference positive control drug Temozolomide (TMZ) (IC50 8.431 ± 4 µg/mL). The results showed that treatment greatly reduced the viability of treated malignant cells. Statistical analysis was performed to determine the IC50 values. Citronellol and CN@AgNPs cytotoxically affected the SF-767 cancer cell line with IC50 of 20.04 ± 4 µg/mL and 19.67 µg/mL (Fig. 19).The significantly higher efficacy of the nanoformulation suggests that CN@AgNPs may offer potential benefits in terms of transport and long-lasting anticancer action (Fig. 19).
SF-767 glioblastoma cell line viability after treatment citronellol (green) and CN@AgNPs (yellow) was assessed using the MTT assay and results are presented as mean ± 95% confidence interval (CI) with error bars. Statistical significance was determined using two-way ANOVA followed by Tukey’s post hoc analysis, with significance levels indicated as ns (not significant), p < 0.05, p < 0.01, p < 0.001, and p < 0.0001.
Use of IC50 values to parameterize concentration–effect curves. A plot of the concentration–response curve for Citronellol (A). A plot of the concentration–response curve for CN@AgNPs (B), and Temozolomide (Positive control) (C).
Effect of Compounds on P53 and CDK4 expression in glioblastoma (SF-767) using real-time polymerase chain reaction (RT-PCR).
The β-actin gene was used as a reference in RT-PCR analysis to measure gene expression in the cells following a 24-hour exposure to the IC50 concentrations of the drugs. The results showed that in comparison to the untreated cells, the glioblastoma cell lines treated with chemicals had downregulated CDK4 and up-regulated P53 levels. Furthermore, the CN@AgNPs had a noticeably stronger effect than free citronellol (Fig. 20).
Relative gene expression levels of p53 (green) and CDK4 (red) in cells treated with citronellol and CN@AgNPs, analyzed using RT-PCR. Data are presented as mean ± 95% confidence interval. Statistical significance was assessed using a two-way ANOVA, followed by a Tukey’s post hoc test to compare each treatment group with the control.
Discussion
In recent years, plant-derived natural agents have emerged as a key area of focus for cancer management65. Numerous plant extracts have been tested for their ability to prevent cancer and have been demonstrated to exert cytotoxic or cytostatic effects on a range of cancer cell lines65. Such research is thought to be beneficial in providing scientific data for the development of a powerful medication that can be used to treat a variety of tumors without causing any negative side effects. Citronellol, a monoterpene, was used in this investigation15. The bioavailability and targeted delivery of therapeutic drugs within the bulk of GBM tumors can now be enhanced by loading them onto nanoparticles (NPs)18. Because silver nanoparticles (AgNPs) have superior catalytic optical and biological capabilities, they are among the most widely produced, studied, and used examples of such materials66. For instance, citronellol-based silver nanoconjugates (CN@AgNPs) were prepared and their cytotoxic effects on glioblastoma were examined.
Citronellol-CMC was utilized as a biocomposite in conjunction with AgNPs to produce nanoformulations. After adding NaOH to reduce the pH to a basic level, the synthesis was completed when the color changed from colorless to pale brown, indicating the formation of Ag NPs. Ag + changed into metallic Ag (Ag) nanoparticles when the solution was neutralized by hydroxyl ions (OH −) from NaOH. The ability of antioxidants to decrease and stabilize is enhanced when Ag-NPs are produced from olive leaf extract in an alkaline environment (Khalil et al., 2014)67. In contrast, the nanoparticles generated in low-pH (2, 4, and 6) solutions are incredibly large and unstable68. Zeta size, zeta potential, UV-vis, FTIR, XRD, and SEM were used to characterize the synthesized silver nanoconjugates (Figs. 2, 3, 4 and 5).
The stability of the colloidal dispersion, free from agglomeration or precipitate formation, was ensured by PDI values less than 0.569. The acquired PDI value showed that all the silver nanoparticles in the solutions had a uniform size distribution and did not exhibit any significant differences (p > 0.05) over the course of the temperature and storage time evaluation. The capping of biomolecules caused the nanoparticles to be nearly stable despite their decreased zeta potential value, and the combined effect of the capping agent and the nanoparticles was responsible for the particle zeta potential value. In addition, the capping ingredient provides nanoparticle stability by preventing agglomeration70,71. One of the most significant and simple methods for verifying the formation of nanoparticles is UV-Vis spectroscopy71. According to the scanning electron microscopy results, the solitary absorption peak in the new study indicates the presence of spherical nanoparticles. The CN@AgNP approach was used to determine the size of AgNPs, which was in line with earlier findings72. The stretching vibration of the O-H bond arising from the group present in flavonoids, terpenoids, saponins, and polyphenol compounds is the characteristic of the absorption band of citronella oil at 3391 cm−1, which is consistent with the FTIR spectra, according to Mulwandari et al.73. The stretching vibration of the C–O alcohol bond at 1013 cm−1, which shows a significant absorption capacity, supports this notion. Furthermore, the absorption at 1722 cm−1 indicates the presence of an aromatic C = C bond vibration. Nevertheless, the absorption strength in the Ag nanoparticles’ (AgNPs’) wavenumber region remained unchanged before synthesis. It did, however, exhibit a wavenumber shift, indicating that Ag and the OH functional group interacted to produce EO-AgNPs (3395–3422 cm−1). The synthesized AgNPs’ XRD analysis confirmed the previously reported face-centered cubic (FCC) crystal structure of Ag74,75,76.
Target proteins that bind tiny compounds of interest can be identified directly using a variety of biochemical affinity purification techniques33,77. These elucidation tests are expensive and time-consuming, even though they can clearly identify compound-target interactions. Using existing bioactivity data, the well-established computer method known as “in silico protein target prediction” provides an alternate method for determining target-ligand interactions78. In the fields of toxicity and efficacy prediction, these techniques are crucial33. These methods are intended to identify orphan chemical targets at an early stage of drug development, and predictions will thereafter serve as the foundation for experimental confirmation. Protein targets of small-molecule ligands can be predicted using both structure-based and ligand-based approaches.
Citronellol was recognized as a possible p53 ligand by the chemogenomics program PIDGINv2. The most promising potential targets for ligand interactions were identified as CDK4 and p53 by additional research using the STRING and KEGG databases. To verify the interaction of citronellol and CN@AgNPs with p53 (PDB ID: 8DC8) and CDK4 (AF-P11802-F1-v4), an in silico docking study was conducted in the current work (Tables 4 and 5). Temozolomide, A FDA approved glioblastoma drug, was also included for comparison with the selected ligands. Two major mechanistic strategies have been used among the compounds created to target p53. Blocking the p53/MDM2 complex and focusing on wild-type p53 is one way to prevent its degradation. Targeting p53 mutations was the second most common mutation. Numerous treatment strategies use small-molecule drugs to restore mutant p53 function79. Comparing the P53 docking contact with free citronellol and CN@AgNPs, it was shown that the stability of the interaction was higher in CN@AgNPs than in its free ligand and reference drug. At the hydrophobic pocket created by Y220C, CN@AgNPs formed a metallic and hydrophobic connection with C220. Y220C is a relatively specific p53 mutation. The conversion of tyrosine to cysteine between S7 and S8 results in a hydrophobic pocket on the p53 protein surface. However, Y220 is positioned far from the site where p53 interacts with the DNA80. Thus, Y220C is an ideal therapeutic target because it targets its hydrophobic pocket, which helps p53 restore its normal folding without obstructing its ability to bind to DNA.466 It has been shown that several small compounds can bind to Y220C’s hydrophobic pocket and restore p53 function79. The binding pattern and interaction of the citronellol complex with p53 and CDK4 were further evaluated using MD simulations. RMSF, Rg plots, DCCM, and PCA analysis provide crucial information for characterizing local changes along the protein chain81.
The ATP-binding region of CDK4 comprises 93–102 residues. Drug designs targeting the same pocket have been described by several groups82. When the docking interaction between CDK4 with reference drug, CN@AgNPs and free citronellol at ATP-binding site was examined, CDK4 showed the comparable binding affinity for both free citronellol and CN@AgNPs, with docking scores of -5.6 and − 5.8 kcal/mol, respectively. Hydrophobic interactions with CDK4 residues, metal acceptor bonds (CN@AgNPs), and hydrogen bonds were observed. These findings are compatible with those of another study83, which showed that alkyl and pi-alkyl with V20, A157, and L35 exhibited an inhibitory binding mechanism and that H-bond contact with D158 and V96 was consistent with the present study.
Citronellol and CN@AgNPs were evaluated for their cytotoxic effects on the human glioblastoma cell line SF-767 using an in vitro MTT assay. The IC₅₀ values of citronellol (20.04 ± 4 µg/mL) and CN@AgNPs (19.67 µg/mL) indicated anti-tumor activity that was comparable to that of temozolomide against SF-767 cells (Fig. 19). The primary cause of the inability to attain therapeutic medication concentrations in the brain tissue has been attributed to the impermeable nature of the blood-brain barrier (BBB). However, employing nanoparticles can enhance drug release efficiency, reduce adverse drug reactions, and improve treatment outcomes84,85,86. Compared with the cell lines studied, AgNPs showed extremely effective selective cytotoxic action against glioma cells. Our results are in line with earlier research, and the generated nanomaterials can greatly enhance glioblastoma therapy.
The tumor suppressor protein p53 is essential for preventing cancer in healthy cells by inducing cell cycle arrest, apoptosis, and DNA repair87. One of the main targets of p53 in the regulation of the cell cycle is the cyclin-dependent kinase CDK inhibitor p21. Cell cycle arrest occurs when the cyclin D1/CDK4 complex is suppressed or attaches to the G1 phase65. As a result, it significantly affects the cell cycle regulation. Targeting the genes necessary for cell cycle arrest or death, p53 is activated by a variety of intrinsic and external stress signals65. Therefore, the ability of the investigated drugs to increase P53 can aid in triggering apoptosis. It is important to note that malignancies in patients with altered P53 activity respond better to novel genotoxic chemotherapeutics that function through the P53 pathway than to conservative chemotherapeutic medications83. A small chemical causes p21 expression in both wt-p53 and mut-p53 cells, stabilizes p53 conformation, and increases p53 activity in GBM cells88.
Furthermore, the cells were exposed to IC50 concentrations of the samples for 24 h prior to RT-PCR for gene expression analysis to validate these results for p53 and CDK4 expression in glioblastoma (Fig. 20). According to our findings, both substances increased the expression levels of p53 and decreased CDK4 expression in treated cancer cells. The effectiveness of CN@AgNPs was greater than that of citronellol.
These results prompted us to develop a treatment that is clinically feasible in more detail. To optimize therapeutic efficacy, the following phases will concentrate on in vivo validation, thorough mechanistic investigations, and investigation of the possibilities of combination therapy.
Conclusion
This work demonstrates the potential of citronellol-conjugated AgNPs (CN@AgNPs) and free citronellol as an effective therapeutic strategy for the treatment of glioblastoma. Effective drug administration across the blood-brain barrier (BBB) is made possible by the small size of CN@AgNPs, their biocompatibility, improved stability, and low toxicity. Compared to free citronellol, CN@AgNPs showed greater binding affinities and stable interactions with CDK4 and p53 active sites, as evidenced by the creation of metallic bonds, according to molecular docking studies. In vitro tests further supported these interactions, showing that CN@AgNPs significantly affected glioblastoma cells. Mechanistically, CN@AgNPs increased the expression of the tumor suppressor gene p53 and decreased that of CDK4, indicating that they can alter important cell cycle regulators. These results, when coupled with the decreased cell viability observed in MTT experiments, suggest a possible involvement in preventing the growth of cancer cells and potentially causing cell cycle arrest, which calls for further research. Our study highlights the potential of CN@AgNPs as a cutting-edge nanotherapeutic platform for glioblastoma targeting and lays the groundwork for further preclinical and clinical research.
Data availability
The dataset analyzed during the current study are available in the manuscript and its supplementary files.
References
Louis, D. N. et al. The 2021 WHO classification of tumors of the central nervous system: a summary. Neuro-oncology 23 (8), 1231–1251 (2021).
Banerjee, K. et al. Current approaches for glioma gene therapy and virotherapy. Front. Mol. Neurosci. 14, 621831 (2021).
Wang, H. H. et al. GADD45A plays a protective role against Temozolomide treatment in glioblastoma cells. Sci. Rep. 7 (1), 8814 (2017).
Mowforth, O. D. et al. Personalised therapeutic approaches to glioblastoma: A systematic review. Front. Med. 10, 1166104 (2023).
Wu, W. et al. Glioblastoma multiforme (GBM): an overview of current therapies and mechanisms of resistance. Pharmacol. Res. 171, 105780 (2021).
Mo, F. et al. Blood–brain barrier in brain tumors: biology and clinical relevance. Int. J. Mol. Sci. 22 (23), 12654 (2021).
Majchrzak-Celińska, A. & Studzińska-Sroka, E. New avenues and major achievements in phytocompounds research for glioblastoma therapy. Molecules 29 (7), 1682 (2024).
Freitag, T. et al. Combined Inhibition of EZH2 and CDK4/6 perturbs Endoplasmic reticulum-mitochondrial homeostasis and increases antitumor activity against glioblastoma. NPJ Precision Oncol. 8 (1), 156 (2024).
Koutsodontis, G. et al. Sp1 plays a critical role in the transcriptional activation of the human cyclin-dependent kinase inhibitor p21WAF1/Cip1 gene by the p53 tumor suppressor protein. J. Biol. Chem. 276 (31), 29116–29125 (2001).
Merlin, J. J. et al. Role of dietary antioxidants in p53-mediated cancer chemoprevention and tumor suppression. Oxidative Med. Cell. Longev. 2021 (1), 9924328 (2021).
Xie, C. et al. Persimmon (Diospyros Kaki L.) leaves: a review on traditional uses, phytochemistry and Pharmacological properties. J. Ethnopharmacol. 163, 229–240 (2015).
Fatima, K. & Luqman, S. Citronellal suppress the activity of ornithine decarboxylase in hypopharyngeal carcinoma cells. South. Afr. J. Bot. 143, 443–448 (2021).
Fatima, M. et al. Reactivity and Structure Analysis of Anticancer Properties of Citronellol, Linalool and Geraniol Essential Oils by Using DFT Calculation. Linalool Geraniol Essential Oils by Using DFT Calculation.
Bicak, B. et al. Structural analyses (Spectroscopic and in silico), DNA-binding and in vitro studies of two popular Sage species (Salvia officinalis and salvia fruticosa). J. Essent. Oil Bearing Plants. 28 (1), 207–223 (2025).
Yu, W. N. et al. Citronellol induces necroptosis of human lung cancer cells via TNF-α pathway and reactive oxygen species accumulation. Vivo 33 (4), 1193–1201 (2019).
Zhuang, S. R. et al. Effect of citronellol and the Chinese medical herb complex on cellular immunity of cancer patients receiving chemotherapy/radiotherapy. Phytotherapy Research: Int. J. Devoted Pharmacol. Toxicol. Evaluation Nat. Prod. Derivatives. 23 (6), 785–790 (2009).
Ho, Y. et al. Anticancer effect of citrus hystrix DC. leaf extract and its bioactive constituents citronellol and, citronellal on the triple negative breast cancer MDA-MB-231 cell line. Pharmaceuticals 13 (12), 476 (2020).
Pistollato, F. et al. Targeting glioblastoma with the use of phytocompounds and nanoparticles. Target. Oncol. 11, 1–16 (2016).
Bose, A. et al. Nanomicelles: types, properties and applications in drug delivery. IET Nanobiotechnol. 15 (1), 19–27 (2021).
Li, L. et al. Application of nanomicelles in enhancing bioavailability and biological efficacy of bioactive nutrients. Polymers 14 (16), 3278 (2022).
Akhtar, N. et al. Fabrication of ethosomes containing Tocopherol acetate to enhance transdermal permeation: in vitro and ex vivo characterizations. Gels. 8(6), 335 (2022).
Al-Serwi, R. H. et al. Targeted drug administration onto cancer cells using hyaluronic acid–quercetin-conjugated silver nanoparticles. Molecules 28 (10), 4146 (2023).
Mohamed, J. M. M. et al. Polymeric ethosomal gel loaded with nimodipine: optimisation, Pharmacokinetic and histopathological analysis. Saudi Pharm. J. 30 (11), 1603–1611 (2022).
Asif, M. et al. Green synthesis of silver nanoparticles (AgNPs), structural characterization, and their antibacterial potential. Dose-Response 20 (2), 15593258221088709 (2022).
Wang, D. et al. Fungus-mediated green synthesis of nano-silver using Aspergillus sydowii and its antifungal/antiproliferative activities. Sci. Rep. 11 (1), 10356 (2021).
Gherasim, O. et al. Biodistribution of essential oil-conjugated silver nanoparticles. Rom J. Morphol. Embryol. 61 (4), 1099–1109 (2020).
Nasiriboroumand, M., Montazer, M. & Barani, H. Preparation and characterization of biocompatible silver nanoparticles using pomegranate Peel extract. J. Photochem. Photobiology B: Biology. 179, 98–104 (2018).
Ejaz, S. A. et al. In-silico investigations of quinine and Quinidine as potential inhibitors of AKR1B1 and AKR1B10: functional and structural characterization. Plos One. 17 (10), e0271602 (2022).
Ullah, W. et al. Dual Therapeutic Potential of Novel Cornallegic Acid: in Vitro Anti-Inflammatory and ROS Suppression with Computational insights from Docking, ADMET, and DFT Studies (Journal of Computational Biophysics Chemistry, 2025).
Daina, A., Michielin, O. & Zoete, V. SwissADME: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci. Rep. 7 (1), 42717 (2017).
Lynch, T. & Price, A. The effect of cytochrome P450 metabolism on drug response, interactions, and adverse effects. Am. Family Phys. 76 (3), 391–396 (2007).
Borba, J. V. et al. STopTox: an in Silico alternative to animal testing for acute systemic and topical toxicity. Environ. Health Perspect. 130 (2), 027012 (2022).
Mervin, L. H. et al. Target prediction utilising negative bioactivity data covering large chemical space. J. Cheminform. 7, 1–16 (2015).
Mervin, L. H. et al. Extending in Silico protein target prediction models to include functional effects. Front. Pharmacol. 9, 613 (2018).
RDKit: Open-Source Cheminformatics Software. [cited 2015 7 September]. (2015).
Wong, K. K. et al. Towards the mode of action of strobilanthes Crispus through integrated computational and experimental analyses. J. Plant. Biochem. Biotechnol. 26, 451–466 (2017).
Antunes, S. S., Rabelo, V. W. H. & Romeiro, N. C. Natural products from Brazilian biodiversity identified as potential inhibitors of PknA and PknB of M. tuberculosis using molecular modeling tools. Computers Biology Med. 136, 104694 (2021).
Hengphasatporn, K. et al. Target identification using homopharma and network-based methods for predicting compounds against dengue virus-infected cells. Molecules 25 (8), 1883 (2020).
Wu, Y. et al. A network Pharmacology approach to explore active compounds and Pharmacological mechanisms of a patented Chinese herbal medicine in the treatment of endometriosis. PloS One. 17 (2), e0263614 (2022).
Tumuluru, P. & Ravi, B. Dijkstra’s based identification of lung cancer related genes using PPI networks. Int. J. Comput. Appl. 975, 8887 (2017).
Szklarczyk, D. et al. The STRING database in 2017: quality-controlled protein–protein association networks, made broadly accessible. Nucleic Acids Res. 45 (D1), D362–D368 (2016).
Kanehisa, M. et al. KEGG: biological systems database as a model of the real world. Nucleic Acids Res. 53 (D1), D672–D677 (2025).
Kanehisa, M. Toward Understanding the origin and evolution of cellular organisms. Protein Sci. 28 (11), 1947–1951 (2019).
Kanehisa, M. & Goto, S. KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 28 (1), 27–30 (2000).
Zaelani, B. F., Safithri, M. & Andrianto, D. Molecular Docking of red betel (Piper crocatum Ruiz & Pav) bioactive compounds as HMG-CoA reductase inhibitor. J. Sci. Appl. Chem. 24 (3), 101–107 (2021).
Eberhardt, J. et al. AutoDock Vina 1.2.0: new Docking methods, expanded force field, and python bindings. J. Chem. Inf. Model. 61 (8), 3891–3898 (2021).
Sakhawat, A. et al. Natural compound targeting BDNF V66M variant: insights from in Silico Docking and molecular analysis. Amb Express. 13 (1), 134 (2023).
Kumar, A. et al. Molecular modeling, Docking and dynamics studies of Fenugreek (Trigonella foenum-graecum) α-amylase. J. Biomol. Struct. Dynamics. 41 (19), 9297–9312 (2023).
O’Boyle, N. M. et al. Open babel: an open chemical toolbox. J. Cheminform. 3, 33 (2011).
Rauf, A., Khan, M. & Akram, Z. Lipoxygenase inhibitory potential of secondary metabolites isolated from pistacia integerrima: a comprehensive in vitro analysis integrating molecular docking, ADMET and DFT studies. Tradit Med. Res. 10 (9), 52 (2025).
Shi, D. et al. Computational insight into the small molecule intervening PD-L1 dimerization and the potential structure-activity relationship. Front. Chenistry. 7, 764 (2019).
BIOVIA DS. BIOVIA discovery studio visualizer. 20: p. 779. (2016).
Jini, D. et al. In vitro and in Silico studies of silver nanoparticles (AgNPs) from allium sativum against diabetes. Sci. Rep. 12 (1), 22109 (2022).
Islam, S. et al. Antibacterial potential of propolis: molecular docking, simulation and toxicity analysis. AMB Express. 14 (1), 81 (2024).
Habib, I. et al. Integrated computational approaches for designing potent pyrimidine-based CDK9 inhibitors: 3D-QSAR, docking, and molecular dynamics simulations. Comput. Biology Chem. 108, 108003 (2024).
Ahmad, I. et al. Identification of novel amides and alkaloids as putative inhibitors of dopamine transporter for schizophrenia using computer-aided virtual screening. Front. Pharmacol. 16, 1509263 (2025).
Arantes, P. R. et al. Making it rain: cloud-based molecular simulations for everyone. J. Chem. Inform. Model. 61 (10), 4852–4856 (2021).
Chen, S. F. et al. Binding selectivity studies of PKBα using molecular dynamics simulation and free energy calculations. J. Mol. Model. 19 (11), 5097–5112 (2013).
Cheng, Y., Dai, C. & Zhang, J. SIRT3-SOD2-ROS pathway is involved in linalool-induced glioma cell apoptotic death. Acta Biochim. Pol. 64 (2), 343–350 (2017).
Osanloo, M. et al. Comparison effects of ferula gummosa essential oil and Beta-pinene alginate nanoparticles on human melanoma and breast cancer cells proliferation and apoptotic index in short term Normobaric hyperoxic model. BMC Complement. Med. Ther. 23 (1), 428 (2023).
Mirabdaly, S. et al. Effects of Temozolomide on U87MG glioblastoma cell expression of CXCR4, MMP2, MMP9, VEGF, anti-proliferatory cytotoxic and apoptotic properties. Mol. Biol. Rep. 47, 1187–1197 (2020).
Danaei, M. et al. Impact of particle size and polydispersity index on the clinical applications of lipidic nanocarrier systems. Pharmaceutics. 10(2), 57 (2018).
Fadaka, A. O. et al. Broad spectrum Anti-Bacterial activity and Non-Selective toxicity of gum Arabic silver nanoparticles. Int. J. Mol. Sci., 23(3), 1–18 (2022).
ASTM Data File No. 04-0783.
Shakya, G., Balasubramanian, S. & Rajagopalan, R. Methanol extract of wheatgrass induces G1 cell cycle arrest in a p53-dependent manner and down regulates the expression of Cyclin D1 in human laryngeal cancer cells-an in vitro and in Silico approach. Pharmacognosy Magazine. 11 (Suppl 1), S139 (2015).
Pryshchepa, O., Pomastowski, P. & Buszewski, B. Silver nanoparticles: synthesis, investigation techniques, and properties. Adv. Colloid Interface Sci. 284, 102246 (2020).
Khalil, M. M. et al. Green synthesis of silver nanoparticles using Olive leaf extract and its antibacterial activity. Arab. J. Chem. 7 (6), 1131–1139 (2014).
Velgosová, O., Mražíková, A. & Marcinčáková, R. Influence of pH on green synthesis of ag nanoparticles. Mater. Lett. 180, 336–339 (2016).
Maciel, M. V. O. B. et al. Syzygium aromaticum L.(clove) essential oil as a reducing agent for the green synthesis of silver nanoparticles. Open. J. Appl. Sci. 9 (2), 45–54 (2019).
Das, P. et al. Potential therapeutic activity of Phlogacanthus thyrsiformis Hardow (Mabb) flower extract and its biofabricated silver nanoparticles against chemically induced urolithiasis in male Wistar rats. Int. J. Biol. Macromol. 103, 621–629 (2017).
Dighade, R. et al. Nephroprotective effect of bryophyllum pinnatum-mediated silver nanoparticles in ethylene glycol‐induced urolithiasis in rat. IET Nanobiotechnol. 15 (3), 266–276 (2021).
Gaikwad, S. et al. Antiviral activity of mycosynthesized silver nanoparticles against herpes simplex virus and human parainfluenza virus type 3. Int. J. Nanomed. 8 (1), 4303–4314 (2013).
Mulwandari, M. et al. Direct synthesis of Lemongrass (Cymbopogon citratus L.) essential oil-silver nanoparticles (EO-AgNPs) as biopesticides and application for lichen Inhibition on stones. Heliyon. 8(6), 1–11 (2022).
Neelgund, G. M. et al. Single-step, size-controlled synthesis of colloidal silver nanoparticles stabilized by Octadecylamine. Appl. Surf. Sci. 356, 726–731 (2015).
Feng, H. L. et al. Study on the crystalline structure and the thermal stability of silver-oxide films deposited by using direct-current reactive Magnetron sputtering methods. J. Korean Phys. Soc., 56 (1), 1176–1179 (2010).
Sathishkumar, M. et al. Cinnamon zeylanicum bark extract and powder mediated green synthesis of nano-crystalline silver particles and its bactericidal activity. Colloids Surf. B: Biointerfaces. 73 (2), 332–338 (2009).
Rix, U. & Superti-Furga, G. Target profiling of small molecules by chemical proteomics. Nat. Chem. Biol. 5 (9), 616–624 (2009).
Koutsoukas, A. et al. From in Silico target prediction to multi-target drug design: current databases, methods and applications. J. Proteom. 74 (12), 2554–2574 (2011).
Wang, H. et al. Targeting p53 pathways: mechanisms, structures, and advances in therapy. Signal. Transduct. Target. Therapy. 8 (1), 92 (2023).
Bykov, V. J. & Wiman, K. G. Mutant p53 reactivation by small molecules makes its way to the clinic. FEBS Lett. 588 (16), 2622–2627 (2014).
Kumar, G. S. et al. Computational investigations on the natural small molecule as an inhibitor of programmed death ligand 1 for cancer immunotherapy. Life 12 (5), 659 (2022).
Sivashanmugam, M., Raghunath, C. & Vetrivel, U. Virtual screening studies reveal Linarin as a potential natural inhibitor targeting CDK4 in retinoblastoma. J. Pharmacol. Pharmacotherapeutics. 4 (4), 256–264 (2013).
Abo-Salem, H. M. et al. Synthesis, bioactivity assessment, molecular Docking and ADMET studies of new Chromone congeners exhibiting potent anticancer activity. Sci. Rep. 14 (1), 9636 (2024).
Tunç, T. Synthesis and characterization of silver nanoparticles loaded with carboplatin as a potential antimicrobial and cancer therapy. Cancer Nanotechnol. 15 (1), 2 (2024).
Giakoumettis, D., Kritis, A. & Foroglou, N. C6 cell line: the gold standard in glioma research. Hippokratia 22 (3), 105 (2018).
Ghaferi, M. et al. Impact of pegylated liposomal doxorubicin and carboplatin combination on glioblastoma. Pharmaceutics 14 (10), 2183 (2022).
Shyur, L. F. et al. Taiwanin A inhibits MCF-7 cancer cell activity through induction of oxidative stress, upregulation of DNA damage checkpoint kinases, and activation of p53 and fasl/fas signaling pathways. Phytomedicine 18 (1), 16–24 (2010).
Zhang, Y. et al. The p53 pathway in glioblastoma. Cancers 10 (9), 297 (2018).
Acknowledgements
The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University for funding this work through a Large group Research Project under grant number RGP2/259/46.
Funding
Large Research Project grant number RGP2/259/46, King Khalid University.
Author information
Authors and Affiliations
Contributions
HM and SK carried out research work and wrote the initial draft of the manuscript. MIU, HBG, AAMA, NH and ASA conducted the data analysis. MUK planned and supervised the study and edited the final version of the manuscript. MUK, NH and SK technically reviewed, revised and finalized the draft. All authors reviewed the final version of the manuscript and approved it for publication.
Corresponding author
Ethics declarations
Ethics approval and consent to participate
The institutional ethical committee of University of Lahore approved this research project.
Consent for publication
Not Applicable.
Competing interests
The authors declare no competing interests.
Data transparency
Rest assured, I have ensured that all data, materials, software applications, and custom code supporting the claims made in this article are in full compliance with field standards. It is important to note that I have taken into account the possibility of individual journal policies regarding research data sharing, considering the norms and expectations of our discipline.
Clinical trial number
Not applicable.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Manzoor, H., Khan, M.U., Khan, S. et al. Citronellol silver nanoconjugates as a therapeutic strategy for glioblastoma through computational and experimental evaluation. Sci Rep 15, 34076 (2025). https://doi.org/10.1038/s41598-025-14557-0
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-025-14557-0
