Unveiling the anti-glioma potential of a marine derivative, Fucoidan: its synergistic cytotoxicity with Temozolomide—an in vitro and in silico experimental study

Abstract

Glioma coined as a “butterfly” tumor associated with a dismal prognosis. Marine algal compounds with the richest sources of bioactive components act as significant anti-tumor therapeutics. However, there is a paucity of studies conducted on Fucoidan to enhance the anti-glioma efficacy of Temozolomide. Therefore, the present study aimed to evaluate the synergistic anti-proliferative, anti-inflammatory and pro-apoptotic effects of Fucoidan with Temozolomide in in vitro and in silico experimental setup. The anti-proliferative effects of Temozolomide and Fucoidan were evaluated on C6 glioma cells by MTT and migration assay. Modulation of inflammatory markers and apoptosis induction was affirmed at the morphological and transcriptional level by dual staining and gene expression. Molecular docking (MD) and molecular dynamics simulation (MDS) studies were performed against the targets to rationalize the inhibitory effect. The dual-drug combination significantly reduced the cell viability and migration of glioma cells in a synergistic dose-dependent manner. At the molecular level, the dual-drug combination significantly down-regulated inflammatory genes with a concomitant upregulation of pro-apoptotic marker. In consensus with our in vitro findings, molecular docking and simulation studies revealed that the anti-tumor ligands: Temozolomide, Fucoidan with 5-(3-Methy1-trizeno)-imidazole-4-carboxamide (MTIC), and 4-amino-5-imidazole-carboxamide (AIC) had the potency to bind to the inflammatory proteins at their active sites, mediated by H-bonds and other non-covalent interactions. The dual-drug combinatorial treatment synergistically inhibited the proliferation, migration of glioma cells and promoted apoptosis; conversely with the down-regulation of inflammatory genes. However, pre-clinical experimental evidence is warranted for the possible translation of this combination.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13205-023-03814-6.

Introduction

Glioma, a primary brain tumor arising from multi-step tumorigenesis of glial cells, is often characterized by poor prognosis with a median 5-years survival rate of less than 10% (Hanif et al. 2017). The standard-of-care treatment for glioma comprises maximal surgical resection of the tumor mass (20 weeks of survival), followed by radiotherapy (36 weeks of survival by radiation and surgery) and chemotherapy (40–50 week survival) employing temozolomide (TMZ) (Weller et al. 2021). Despite such multimodal approach, the prognosis remains dismal and tumor recurrence is inevitable. Thus, there is an urgent need for alternate therapies against glioma.

The standard drug discovery pipeline is frequently constrained by high expense and low success burden, which is not surprising given the ongoing hunt for alternative chemotherapeutic medicines to address the aggressiveness of gliomas. In this scenario, repurposing drugs would represent an attractive and effective strategy to improve the odds of new drug development success for glioma; as the safety and pharmacological profiles of the repurposed drugs are well-known, they can be streamlined for their passage through FDA more quickly, besides cost-reduction (Cha et al. 2018). Interestingly, several repurposed drugs are currently under clinical investigation against glioma, thereby highlighting the efficiency of this systemic approach.

While choosing the potent drug candidate for repurposing, epidemiological reports have stated that combining natural compounds with conventional radio and chemotherapy enhances the sensitivity of the standard drug and alleviates therapy-associated complications in various cancers (Mokhtari et al. 2017). Marine natural products (MNPs) are among them and have shown to be efficient biological modulators. The main biological effects that have been reported for these MNPs are their powerful antibacterial, anti-infective, anti-inflammatory, and anti-tumor activities (Karthikeyan et al. 2022). As a result, using MNPs as a lead molecule will hasten the creation of novel anti-cancer medications that are more effective and have fewer adverse effects.

One such MNP is fucoidan (FU), a fucose-containing sulfated polysaccharide derived from brown seaweeds. In various regions of Northern Europe and Asia, the crude extracts of FU are available for purchase as dietary supplements. In recent times, the anti-tumor effect of FU has been studied intensively in colon (Kim and Nam 2018), prostate, breast cancers (Xue et al. 2012) and glioma (Do et al. 2010). Furthermore, the combination of FU with cyclophosphamide radically reduced metastasis in the C57BL/6 Lewis lung cancer mouse model (Jin et al. 2021). Nonetheless, the chemo preventive capabilities of FU in glioma have not yet been investigated. As a single therapeutic drug, FU has been shown till date to improve epigenetic differentiation by inhibiting tumour growth in glioma cells (Liao et al. 2019). Although the effects of FU as a monotherapeutic agent in glioma have been documented, their effectiveness in combination therapy with TMZ and its underlying molecular mechanism have not yet been investigated.

In this lacuna, the current study was designed to gain novel insights on the anti-tumor and pro-apoptotic efficacy of FU, both alone and in combination with TMZ against C6 glioma cells. In addition, to predict and identify the mechanism of cytotoxicity induced by the ligands—TMZ and FU, molecular docking and simulation were performed.

Materials and methods

Cell culture and chemicals

C6 rat glioma cell line (RRID: CVCL_0194; Passage number 48) and Human embryonic kidney cell line (HEK 293 T; Passage number 10) was obtained from National Centre for Cell Sciences (NCCS), Pune, India. The cells were maintained at 37 °C under 5% CO₂ in nutrient mixture Ham’s F12 media and Dulbecco’s modified Eagle’s medium (DMEM), supplemented with 10% FBS and 1% antibiotic solution (100 U/ml penicillin and 100 μg/ml streptomycin). At the rate of 80% confluency, cells were seeded in 6-well, 24-well, or 96-well plates based on the experiments being performed.

Stock solutions of TMZ and FU were prepared in dimethyl sulfoxide (DMSO) and water. The aliquots were stored at − 20 °C. On the day of the experiments, FU and TMZ were dissolved in the culture medium, in which the final concentration of DMSO was less than 0.1% (v/v).

Cytotoxicity assay

The cytotoxicity of the drugs, either alone or in combination was assessed using MTT on C6 and HEK293T cell lines. Briefly, C6 and HEK293T cells were seeded in a 96-well plate at a density of 5 × 10³ cells/well and incubated for 24 h. Cells were then treated with TMZ and FU at working concentrations of 10, 50, 100, 150, 200, 250, 300 µM. Following, cytotoxicity was assessed 24-h post-treatment. Based on the IC₅₀ values of the individual drug-treated cells, TMZ (IC₂₀: 1.2 µM) was combined with varying concentrations of FU (10, 50, 100, 150, 200, 250, 300 µM). After 24 h of dual-drug treatment, media were discarded completely and 20 µl of 5 mg/ml MTT was added to each well and incubated at 37 °C for 3 h. The purple-coloured formazan crystals formed were dissolved in 180 µl of DMSO. Absorbance was read at 570 nm spectrophotometrically (Molecular Devices Spectra-Max M5, USA). The absorbance value of untreated control cells was fixed as 100%. The experiment was repeated at least three times. Cytotoxicity was evaluated using the following formula:

$$\text{Cytotoxicity}=\left[\left(\text{Optical density}\left\{\text{OD}\right\}\text{of treated cell}-\text{OD of blank}\right)/\left(\text{OD of control}-\text{OD of blank}\right)\times 100\right].$$

Calculation of selectivity index (SI)

To determine the selectivity and specificity of TMZ, FU, and their combination on glioma cells, SI was calculated using the following formula: SI = (IC₅₀ of the individual and dual-drugs on HEK293T)/(IC₅₀ values obtained for the same treatment regimen on C6 glioma cells). A higher SI value indicates more selective and sensitivity towards cancer cells of the probed drugs and their combination.

Combination index analysis, normalized isobologram, and Fa–CI plot

The statistical combinatorial drug index (CI) was calculated to analyse the interaction pattern of the drugs, while the Chou–Talalay method was used to determine the CI value. The CI values were calculated manually using the CI equation: CI = (D)1/(Dx)1 + (D)2/(Dx)2, where (Dx)1 and (Dx)2 represent the dose of drug 1 and drug 2 in a combination which is required to attain same efficacy as that of drug 1 (D1) and drug 2 (D2), when used alone. According to the Chou–Talalay method, the following criteria of drug–drug interactions were considered by CI: synergistic effect, CI < 1; additive effect, CI = 1; antagonist effect of drugs, CI > 1. Furthermore, dose–effect curve, CI, Fa–CI and drug receptor index (DRI) plots were generated using Compusyn.

Wound-healing assay

To determine the effect of the drugs on the inhibition of glioma migration, C6 cells were plated at a density of 4 × 10⁵ in 6-well plates. At 90% confluency, using a 200 µl sterile plastic tip, a scratch was made on the monolayer of cells. The peeled-off cells were washed using PBS, and thereafter, the cells were maintained in a 2% FBS medium. Following, the cells were treated with the indicated concentration of the drugs both individually and in combination for 24 h. Cell scratch spacing was evaluated at 0 and 24 h by calculating the cell mobility using Image J software (Yang et al. 2020).

Haematoxylin and eosin (H&E) staining

To detect the induction of apoptotic changes at the morphological level, H E staining was carried out (Fischer et al. 2008). For this, 3 × 10⁵ of C6 cells were seeded in sterile coverslips in a 30 mm dish. Following 24 h, cells were treated with the respective IC₅₀ values of the drugs for 24 h. Post-incubation, the medium in each dish was discarded, followed by fixation of cells using 4% paraformaldehyde for 10 min at room temperature. H & E stain was then performed and the cells were imaged at 40 × magnification under a light microscope (Zeiss Light Microscope, Germany) and quantified using Image J.

Acridine orange/ethidium bromide (AO/EtBr) dual staining

To further, validate the changes associated with apoptosis at the morphological level, acridine orange/ethidium bromide (AO/EtBr) dual staining was performed. C6 cells were seeded in a 24-well plate at a density of 20 × 10³/well and were then treated with the drugs, both individually and in combination for 24 h at 37 °C. Following incubation, cells were washed with PBS and stained with 100 μg/ml AO and 100 μg/ml EtBr at room temperature for 5 min. Stained cells were subsequently observed and imaged under a fluorescent microscope (Axiovert 40 CFL, Germany) with an excitation wavelength of 300–360 nm. The number of apoptotic cells per field was counted, and the apoptosis rate was calculated as: % apoptotic cells = number of apoptotic cells/total number of cells × 100, following quantification using ImageJ software.

Nuclear staining

To synchronously confirm the apoptosis induction effect of the drugs at the nuclear level, the fluorescent nuclear stain, 4′6, Diamidino-2-Phenyl Indole (DAPI) was employed. To carry out DAPI staining, C6 glioma cells were seeded at a density of 15 × 10³ cells/well and treated with the respective IC₅₀ value of the drugs. After 24-h post-treatment, the cells were washed with 1X PBS followed by fixation in 4% paraformaldehyde for about 10 min at room temperature. After fixation, the cells were permeabilized using Triton X-100 followed by incubation with 1 μg/ml of DAPI in dark. Using a fluorescent microscope (Axiovert 40 CFL, Germany) with a 359 nm excitation wavelength, stained cells were later examined and photographed. Percentage of apoptosis formation was calculated by ImageJ software.

Quantitative real-time PCR

To quantitatively assess the mRNA transcript levels of inflammatory and pro-apoptotic markers, quantitative real-time PCR (qRT-PCR) was performed. Briefly, total RNA from all of the experimental groups was isolated using RNAiso Plus (Takara, Japan). Complementary DNA (cDNA) was synthesized with HiMedia–cDNA Synthesis Kit (HiMedia, India). qRT-PCR was performed using CFX 96 thermocycler (Bio-Rad, Hercules, CA) and SYBR Green (Takara, Japan) to detect the mRNA. The reaction conditions were as follows: initial denaturation: 95 °C for 30 s; followed by 39 cycles of denaturation at 95 °C for 10 s; annealing at a primer specific annealing temperature for 30 s. The relative gene expression was calculated with the 2^−∆∆C_t method, where ∆∆C_t = (C_{t target gene}–C_{t GAPDH}) sample- (C_{t target gene}–C_{t target gene}) control calibrator. The primers were designed using the primer BLAST, which is listed in Table 1.

Table 1.

Primers designed for each gene

Sl. no.	Genes	Forward primer (5′– > 3′)	Reverse primer (5′– > 3′)	Annealing temperature (°C)
1	IL-6	CTGGTCTTCTGGAGTTCCGT	TGGTCTTGGTCCTTAGCCAC	62
2	TLR-4	TCTGCCCTGCCACCATTTAC	GGAAGTACCTCTATGCAGGGAT	64
3	IFN-γ	ATGGATGCTATGGAAGGAAAGAG	CACTTATGTTGTTGCTGATGGC	58
4	STAT3	TGGGTCTGGCTAGACAAT	TCGTTGGTGTCACACAGAT	58
5	NOS2	TCGTTGGTGTCACACAGAT	CGGCTGGACTTCTCACTCTG	60
6	MYD88	CGACGCCTTCATCTGCTACT	ACCATGCGACGACACCTTTT	54
7	Caspase-3	GGCCGACTTCCTGTATGCTT	GGCCGACTTCCTGTATGCTT	60
8	GAPDH	TCTCTGCTCCTCCCTGTTCTA	TACGGCCAAATCCGTTCACA	55

Molecular docking

Target retrieval

Target retrieval was performed for six major proteins of inflammatory pathway, namely IL-6, TLR-4, JAK2, STAT3, IFN-γ, and NOS2 from UniprotKB (http://ca.expasy.org/sprot/). The 3D structure obtained from PDB was used for further analysis. Detailed information about the targets are listed in Table 2.

Table 2.

List of targets retrieved from Uniprot and PDB

Sl. no.	Protein name	Function	PDB ID	Structure resolution (Å)
1	IL-6	Pro-inflammatory cytokine involved in innate immune response	2L3Y	2.16
2	TLR-4	Cooperates with lipopolysaccharide (LPS) to mediate the innate immune response	2Z64	2.84
3	IFN-γ	Belongs to class of interferon family, aids in immunogenic response against foreign antigens, thereby activating effector immune cells and enhancing antigenic presentation	1FYH	2.04
4	JAK2	Involved in various processes such as cell growth, development, and differentiation or histone modifications. Also involved in innate and adaptive immunity	2HDX	2.35
5	STAT3B	Mediates cellular responses to interleukins and also acts as a regulator of inflammatory responses	1BG1	2.25
6	NOS2	Messenger molecule with diverse functions, such as generation of nitric oxide and serves as a secondary messenger. It can mediate tumoricidal activity in macrophages	3NQS	2.20

Target preparation and active site prediction

The protein preparation wizard of the Schrodinger suite was used to construct the three-dimensional (3D) structures of the six targets, which were individually imported into the maestro (Schrödinger Release 2019-4: Maestro, Schrödinger, LLC, New York, 2019). The receptor preparation process includes: modelling of missing residues and missing side chains, the addition of missing hydrogen atoms, correcting the missing atoms in residues, redundancy validation in occupancies of atoms, modification of metal ionization states and assigning proper charges, removal of undesired water molecules and ideal protonation state of histidine. The presence of negatively and positively charged amino acid residues facilitates to interact with the surrounding moieties across the ligand–protein complex, thereby rationalizing the presence of these three residues specifically, namely Arg, Gln and His. These drug target amino acid residues may have been spun to promote hydrogen and other non-bonded interactions between the surrounding moiety and the ligand–protein complexes. The receptors were optimized using OPLS-3e force field and subsequently minimized to relax the bond angles, lengths and associated clashes. After preparation for all the six targets, the binding site information, which is used for the prediction of protein–ligand interactions, was collected using sitemap protocol (Schrodinger suite).

Ligand retrieval and preparation

TMZ, FU from Fucus vesiculosus, 5-(3-Methy1-trizeno)-imidazole-4-carboxamide (MTIC), and 4-amino-5-imidazole-carboxamide (AIC) were the ligands employed for this study.

TMZ being a pro-drug, at physiological pH is rapidly hydrolyzed into the active form of the drug, 5-(3-methyltriazen-1-yl) imidazole-4-carboxamide (MTIC). This is subsequently metabolised to create 4-amino-5-imidazole-carboxamide, the last byproduct of TMZ’s breakdown (AIC). From Pubchem, the 3D structures of the ligands TMZ, FU, MTIC, and AIC were obtained. The retrieved files were converted into.pdb format using Pymol. The LigPrep module was used to prepare each compound once it had been individually imported into Maestro (Schrödinger Version 2019-4: Maestro, Schrödinger, LLC, New York, NY, 2019).

Molecular docking analysis

The Sitemap procedure was used to estimate the binding efficiency of compounds in the active site of targets (Schrodinger suite) using the Glide XP (extra precision) docking approach, which is listed in Supplementary Table 1. A grid box was created using the receptor grid generation process (Schrodinger suite) to designate the docking area in the protein, allowing for more precise ligand binding score. The prepared targets and ligands were given as input for glide docking. Molecular docking simulations were carried out using the default settings. The docking score, glide energy and number of hydrogen bond (HB) contacts between the target and the ligand were used to validate the simulation results.

Molecular dynamics simulation (MDS)

The structural and dynamic changes accompanied with each of the protein–ligand complexes were analysed using molecular dynamic simulation. The simulation was executed with Gromacs 5.1.4 suite with GROMOS96 43a1 force field on LINUX based workstation. Gromacs provides high molecular simulation results and has been considered as a standard reference tool for simulation in comparison to Desmond (Abraham et al. 2015). During the initial stage, system relaxation was performed to eliminate unwanted atomic contacts, which may lead to unstable MD simulation. Each of the protein–ligand complexes was solvated in a cubic simulation box under a Simple Point Charge 216 (SPC 216) water environment. There was a free run for 10 ps under the equilibrium period with a force constant of 1000 kJ mol⁻¹ mol⁻². Following energy minimization, the system was equilibrated with reference temperature of 300 K at constant temperature and volume (NVT). Further, pressure was maintained with reference pressure of 1 bar at constant temperature and constant pressure (NPT). Temperature was controlled by V-rescale, a modified approach for Berendsen temperature coupling method. The minimized structures were subjected to MD simulation for 20,000 picoseconds (ps) without restrictions. Analysis for MD simulation was executed in terms of Root-Mean-Square-Deviation (RMSD), Root-Mean-Square-Fluctuation (RMSF), Solvent Accessible Surface Area (SASA), and Radius of Gyration (Rg).

Statistical analysis

The results were elicited with triplicate value and expressed as mean ± SD. All the data were analysed statistically by Student t test and two-way ANOVA using GraphPad Prism 8 (GraphPad Software, San Diego, CA). A p value < 0.05 was considered statistically significant.

Results

Cytotoxic effects of TMZ and FU on glioma cells

The cells were individually treated with various drug doses, including 10, 50, 100, 150, 200, 250, and 300 µM, by MTT assay, in order to validate the cytotoxic effect of the medicines on glioma cells both singly and in combination. Both of the drugs utilised in this study (TMZ and FU) significantly reduced the proliferation of glioma cells.

TMZ provided the strongest growth-inhibitory impact of the two medicines used in the current study, and it did so in a consistent dose-dependent manner (IC₅₀: 50 ± 0.003 µM). Nonetheless, as seen by its IC₅₀ value (150 µM ± 0.02 µM), FU as a standalone treatment demonstrated only a moderate degree of cytotoxicity (Fig. 1A).

Fig. 1.

Cytotoxicity analysis of TMZ and FU, individually and in combination on A C6 and B HEK293T cells. Data points represent mean ± SD from triplicate

After determining the separate IC₅₀ values for each drugs, we evaluated how the combination therapy affected glioma cells. This was accomplished by varying the concentration of FU while maintaining a constant IC₂₀ of TMZ (which was found to be more powerful based on individual IC₅₀ values). Following a 24-h course of treatment, we found that exposure to a combination of treatments significantly (p < 0.05) sensitised glioma cells compared to the effect shown with mono-therapeutic regimens. Notably, the concentrations of TMZ and FU were lowered to 1.2 µM and 89.04 µM, respectively, when given as a combinatorial treatment (Fig. 1A). Our results demonstrate that FU, even at low doses, can increase cell sensitivity to TMZ. After 24 h of treatment, the IC₅₀ values of TMZ-alone, FU-alone and TMZ + FU in C6 glioma cells were found to be 50 ± 0.003 µM,150 ± 0.02 µM and 89.04 ± 0.05 µM, respectively (Fig. 1A).

Moreover, cytotoxicity was tested on HEK293T, a normally transformed cell line, to determine if TMZ and FU therapy alone or in combination were harmful to normal cells. We observed that HEK293T cells had higher IC₅₀ values for single- and dual-drug treatment (TMZ + FU) than glioma cells. These results imply that TMZ + FU therapy, both individually and in combination, was more cytotoxic and sensitive to glioma cells than to normal cells (Fig. 1B).

Selectivity index (SI)

On the basis of cytotoxicity assay on C6 and HEK293T cells, the selectivity index (SI) of TMZ, FU, and their combination of TMZ + FU was assessed. At concentrations produced below the lethal concentration of normal cells, drugs with higher SI values are thought to be more effective against cancer cell lines. IC₅₀ of TMZ in HEK293T cells was fivefold higher than that of C6 cells, and IC₅₀ of FU in HEK293T cells was twofold higher than that of C6 cells. Nevertheless, the combination of TMZ and FU exhibited a fold change that was 2.24 times higher in HEK293T than that of C6 cells (Table 3).

Table 3.

Selectivity index (SI) analysis

Drugs used	IC50 value in C6 glioma (µM)	IC50 value in HEK293T cells (µM)	Selectivity Index (SI)
TMZ	50	250	5
FU	150	300	2
TMZ + FU	89.04	200	2.24

Synergistic effects of TMZ and FU on glioma cell proliferation

To determine whether the cytotoxic effects of the dual-drug combination (TMZ + FU) were additive, synergistic or antagonistic, normalized isobologram and statistical combination index (SCI) for non-constant ratio combination design was assessed. CI values were calculated by the Compusyn software according to the recommendations of Chou–Talalay, wherein combinatorial-drug treatment (TMZ + FU) exhibited strong synergistic effect at an IC₅₀ value of 89.04 µM (Figs. 1A and 2A), CI index of 0.62 (Fig. 2B). Besides, DRI value along with isobologram analysis of the dual-drug combination was obtained, at its IC₅₀ value, suggesting a favourable synergism (Fig. 2C, D). These results suggest that the combinatorial-drug treatment of TMZ and FU exhibited maximum synergistic cytotoxicity against glioma cells.

Fig. 2.

Combination Index Analysis using Compusyn. A Dose–effect curves for TMZ, FU and TMZ + FU after 24 h of treatment. B Combination index plot: the combination index is plotted as a function of Fa. C Dose reduction index plot for combination: Dose reduction index values at different Fa values for each drug in the combination. D Isobologram analysis: Combination index determined with respect to different drug dosages (A-TMZ and B-FU), and the points represent the CI values obtained below 1

One of the major hallmarks for glioma pathogenesis is migration, which is a key contributing factor for most of the treatment failures. Inhibition of glioma proliferation of the combinatorial treatment drove us to investigate if this dual-drug cocktail could also hamper glioma migration. Thus, we performed wound healing assay and measured the cell-free area at 0 and 24 h, following exposure to the drugs. Though the efficacy of TMZ (Fig. 3F) and FU (Fig. 3G) as individual drugs was relatively modest on sustaining the cell-free area, the combinatorial treatment (TMZ + FU) had halted the migration of glioma cells, as perceived in terms of wound closure area (Fig. 3I, Supplementary Fig. 4). This highlights the fact that the combinatorial treatment (TMZ + FU) besides exhibiting significant anti-proliferative effect, can also inhibit the mobility of glioma cells (Fig. 3H, I).

Fig. 3.

Inhibition of cell migration under the different treatment groups (0 h: A, B, C and D—control, 24 h: E-Control, F-TMZ, G-FU and H-TMZ + FU respectively. I Cell mobility was interpreted in terms of percentage wound closure area, *** indicates p ≤ 0.001, compared with control

Combinatorial effect of TMZ and FU on apoptosis on C6 glioma cells

We analyzed the induction of apoptosis at the morphological level by a few staining techniques to gain insight into the inhibitory mechanism triggered by TMZ, FU, or their combination on glioma cells. This is due to the fact that evasion of apoptosis is a distinct feature of most of the cancer cells including glioma.

First, we performed H&E staining, wherein we found that treatment with TMZ-alone had endowed a moderate number of apoptotic cells (49.6%, Fig. 4B) with shrinking nuclei and the cells had very poor demarcation of the nucleus and cytoplasm, when compared with that of control (untreated, Fig. 4A) cells. FU treatment on the other hand, resulted only a few number apoptotic cells (Fig. 4C), while most of the cells had a clear cytoplasm and nucleus, denoting that FU as an individual drug had moderate apoptosis induction potency but was not as effective as the standard drug, TMZ (33.06%). Interestingly, cells which were exposed to the combinatorial treatment (TMZ + FU) portrayed a large number of apoptotic cells (84.11%) with cell shrinkage (Fig. 4D), and their viability was drastically reduced. In addition, the cells had displayed marginated, condensed and aggregated nuclei, which all are the morphological characteristics of apoptosis.

Fig. 4.

Apoptotic cell death observed using Hematoxylin and Eosin (H &E) stain under different treatment groups on C6 glioma cells A Control, B TMZ, C FU, D TMZ + FU and E Percentage of cells was calculated as total cellular shrinkage and loss of integrated density *** indicates p < 0.001, compared with control

Following affirmation that the combinatorial treatment restrained glioma growth and induced apoptosis by H&E, we were also interested to distinguish the cells at different stages of apoptosis following administration of these drugs. For this purpose, AO/EtBr stain was employed. AO penetrates the live cells or cells in the early stage of apoptosis and stains them yellow–green, while EtBr is only permeable to the late-apoptotic cells with compromised cell membrane, emitting orange–red fluorescence. In accordance with this principle, cells in control were live and healthy and hence stained green (Supplementary Fig. 1A), whereas, individually TMZ-treated cells (Supplementary Fig. 1B) displayed a large number of cells at the early apoptotic stage (24%), emitting yellow–green fluorescence. On the other hand, treatment with FU (Supplementary Fig. 1C) also resulted in a large number of early apoptotic cells and few late apoptotic cells (32.3%), signifying the early apoptotic effect of FU. Interestingly, combined administration of these drugs as a dual-drug treatment resulted in a large number of late-apoptotic cells (45%), emitting orange–red fluorescence (Supplementary Fig. 1D). In addition, a small proportion of the cells displayed an uneven orange–red fluorescence pattern, with increased cell volume, thereby denoting that the cells had undergone necrosis.

Following the analysis of cytoplasmic changes induced by the combinatorial treatment via H&E and AO/EtBr stains, we then employed the nuclear stain, DAPI to evaluate the alterations induced by TMZ, FU and their combination at nuclear level. Following staining, cells assigned as control, depicted a clear and distinct nucleus (Supplementary Fig. 2A). As expected, the nuclear morphology in the treated cells was altered in all the treatment groups employed for our study. Among the individual drugs, treatment with TMZ endowed 70.48% of apoptotic cells with condensed and marginated nuclei (Supplementary Fig. 2B). However, treatment with FU resulted only in 67.28% of apoptotic cells (Supplementary Fig. 2C) with condensed nuclei. The most significant nuclear alterations were apparent in the combinatorial treatment (TMZ + FU), wherein 73.3% of apoptotic cells with highly shrinked nuclei and patterns of nuclear fragmentation were observed (Supplementary Fig. 2D). Taken together, our staining results suggested that the combinatorial treatment was bestowed with the potency to induce apoptosis at both cytoplasmic and nuclear level, besides suppressing glioma proliferation and migration.

Dual-drug combination modulated the gene expression levels of inflammatory and pro-apoptotic markers in C6 glioma cells

Observations from the staining techniques led us to investigate the underlying molecular mechanism of this combinatorial treatment on glioma cells. Hence, we analysed the relative mRNA expression levels of inflammatory and pro-apoptotic genes namely IL-6, TLR-4, MYD88, IFN-γ, STAT3, NOS-2 and caspase-3 by qRT-PCR analysis. It is worth noting that the inflammatory genes IL-6, TLR-4, MYD88, NOS2, STAT3 and IFN-γ were aberrantly expressed in glioma’s and may contribute to tumor progression and chemo-resistance. Treatment with TMZ and FU, as individual drugs, had significantly down-regulated the expression levels of the inflammatory genes while concomitantly increasing the mRNA expression of caspase-3 to a considerable extent. Interestingly, among all of the experimental groups, the combinatorial treatment of TMZ + FU displayed a significant decrease in the fold change of inflammatory genes while up-regulating the fold change of caspase-3 (Fig. 5), thereby denoting that this combination elicited apoptosis by down-regulating the inflammatory signaling cascade.

Fig. 5.

Anti-glioma effects of FU, both alone and in combination with TMZ on the mean fold change (Linear scale) of inflammatory genes A IL-6, TLR4, MYD88, IFN-γ, NOS2, STAT3 and apoptotic gene, B caspase-3 in C6 glioma cells *** indicates p < 0.001 in comparison to control

Molecular docking

Active site prediction

The active sites were predicted using Sitemap prediction tool (Schrodinger) and have been listed in the Supplementary Table 1.

Molecular docking analysis

In order to predict the interaction of the ligands with the key drivers of inflammation in glioma, docking analysis was performed for the following targets namely, IL-6, TLR4, JAK2, STAT3, IFN-γ and NOS2. The detailed interaction profile of the ligands with inflammatory proteins is tabulated in Table 4.

Table 4.

Interaction profile of all the inflammatory targets against the four ligands

Ligands	Interaction type	Functional moieties {}—Interactions with same amino acid residues	Amino acid residues	Binding energy (Kcal/mole)
IL-6
TMZ	H-bonds (4)/salt bridge (1)	NH2+ (at 1st position of tetrazine ring), NH (6th position of tetrazine ring), O−(7th position) of carboxamide) and NH3+ (8th position) of carboxamide in imidazole ring/	Leu90, Leu89, Ile87, Ser186/ Lys91	− 23.953
MTIC	H-bonds (3)	NH2+ (1st position), N+H2 (3rd position), O− (4th position) of carboxamide and NH2 (5th) position of tetrazine ring	Lys91, Leu90, Arg74	− 13.480
AIC	H-bonds (4)	OH (at 1st position), NH2+ (2nd position), {NH2+ (4th position), substituted chain (at 5th position, NH-3rd position)} of imidazole ring	Gly77, Cy78 and Gln85,	− 28.148
FU	H-bonds (2)	OH (4-Hydroxy), O–SO3 (at 5th position) (sulphate) of Fucose ring/	Leu89, Leu90	− 28.902
TLR4
TMZ	H-bonds (3)/salt bridge (3)	{NH+ at 4th position and N+H2 at 6th position, N+H3 at 8th position} of imidazole ring/{NH+ at 4th position and {N+H2 at 6th position}, N+H3 at 8th position}} of imidazole ring	Asp 208, Glu229/Asp180, Asp208 and Glu229	− 27.448
MTIC	H-bonds (3)/salt bridge (2)	NH2+ (at 1st position), {NH2+ (at 3rd position), NH3+ (at 4th position-of imidazolidin)}/NH2+ (at 1st position), NH2+ (at 3rd position) of imidazole ring	Glu229, Asp208/Glu229 and Asp208	− 40.175
AIC	H-bonds (5)/salt bridge (3)	N+H3 (1st), NH2+ (2nd), NH2+ (4th), 5th-substituted chain (1st and 3rd NH-NH) of imidazole ring/ NH2+ (2nd) of imidazole ring	Glu229, Asp208 / Glu229, Asp180, Asp 208	− 25.524
FU	H-bonds (2)/salt bridge (1)	O-SO3− (at 3rd position) and OH (5-hydroxy) of Fucose ring	Arg233, Arg337/ Lys263	− 23.074
JAK2
TMZ	H-bonds (3)/salt bridge (2)	O−(4th position) of carboxamide, NH2+ at 7th and 8th position of imidazole ring – (OH & NH2)/ O−(4th position) of carboxamide and NH2+ at 7th position of imidazole ring	Leu824, Cys905, Asp 908, Asp908/ Arg912 and Asp908	− 32.530
MTIC	H-bonds (7)	N+H2 (at 1st position), N+H2 (3rd position), OH & NH2 (at 4th position) of imidazole ring	Tyr900, Ser 573, Leu583, His510, Tyr900 and Pro902	− 42.885
AIC	H-bonds (7)/salt bridge (1)	NH2 and OH (1st position), NH and NH2+ (2-(3-methyl-2λ4-triazenyl), NH (3rd position) and NH2+ (5th position) of imidazole ring/NH2+ (5th position) of imidazole ring	Asp963, Arg949, Asn826, Leu824/ Asp963	− 33.159
FU	H-bonds (4)	O-SO3− (3rd) of sulphate, OH at 4th and 5th position of Fucose ring	His510, Arg912, Ser903, Pro902	− 29.593
STAT3B
TMZ	H-bonds (3)/salt bridge (1)	NH (at 1st position), NH3+ (at 9th position), of imidazole ring/ O− at 3rd position (oxidanyl group) of imidazole ring	Glu455, His457/ Lys244	− 17.346
MTIC	H-bonds (4)/pi (π)-cation (1)	{NH2 (at 1st position), NH2} (2nd position at imidazolidin-4-yl), NH (5th position)/ NH2+ (at 3rd position) of imidazole ring	Gln247, Val323 and His457/His457	− 28.236
AIC	H-bonds (4)/salt bridge (2)/pi (π)-cation (1)	NH3+ (at 1st position), {NH2+ (4th position), substituted chain (5th position of 1st and 3rd NH-NH)} of imidazole ring /NH2+ at 2nd position of imidazole ring/ NH2+(2nd position of substituted chain)	Glu455, Asp371/ Glu455, Asp371/ His457	− 20.667
FU	H-bonds (2)	OH at 3rd position and O-SO3− (at 5th position) of Fucose ring	Thr456, His457	− 23.658
IFN-γ
TMZ	H-bonds (4)	NH (at 7th position), OH & NH2 (at 9th position) of imidazole ring	Cys178, Glu175, Thr172	− 31.600
MTIC	H-bonds (3)	{NH (4th position), (carboxamide), NH3+ (5th position)}, of imidazole ring	Cys178, Thr172 and Thr172	− 24.403
AIC	H-bonds (7)/salt bridge (1)	NH (at 1st position), NH-NH (at 1st and 3rd position), NH3+(3rd position), NH2+(4th position) of imidazole ring/NH2+(4th position) of imidazole ring	Cys178, Asp176, Asp176, Glu175 and Gln173	− 36.282
FU	H-bonds (3)	OH (3,4-dihydroxy), O = SO3− of Fucose ring	Phe134, Ile152	− 23.932
NOS2
TMZ	H-bonds (2)/pi (π)-cation (2)	{NH2 and OH} (at 8th position)/NH2+ at 7th position of imidazole ring	Asn364, Phe363 and Trp188	− 24.211
MTIC	H-bond (1)/pi (π)-cation (2)	NH2 (at 2nd position)/ NH2+ (4th position) and C-at 5th position of imidazole ring	Asn364/ Trp188 and Phe363	− 20.902
AIC	H-bonds (6)/salt bridge (1)	{NH2 (at 1st position), (1st and 3rd position NH-NH at 2nd position of substituted chain), {N+H2 (3rd position), N+H2 (5th position) and OH (1st)} of imidazole ring/ N+H2 (3rd position) of imidazole ring	Trp366, Glu361/ Glu361	− 33.902
FU	H-bond (1)	O−-SO3 bonded at 2nd position of Fucose ring	Tyr485	− 24.978

Interestingly, we found that, all of the anti-tumor ligands viz., TMZ, MTIC, AIC and FU interacted with IL-6 by H-bonds predominantly. With regard to the binding energies, we observed that, TMZ, MTIC and AIC formed more number of H-bonds and exerted a binding energy of − 23.953, − 13.480 and − 28.148 kcal/mole (Table 4). Although FU exhibited only two H-bonds with IL-6 (Fig. 6), it depicted a highest binding energy of − 28.902 kcal/mole (Table 4). This could be due to the presence of other non-bonding interactions, which could have facilitated the stronger interaction of FU with IL-6 (Fig. 6).

Fig. 6.

Interaction profile of A IL-6, B TLR4, C JAK2 with ligands—TMZ, MTIC, AIC and FU

In the case of TLR4, we observed that TMZ, MTIC, AIC and FU interacted through H-bonds and salt bridge formation (Fig. 6), with a binding energy of − 27.448, − 40.175, − 25.524, and − 23.074 kcal/mole, respectively (Table 4). Among the anti-tumor ligands, MTIC, despite possessing lesser number of H-bonds and salt bridges, depicted the highest binding energy (Fig. 6).

A similar pattern of interaction was observed in the case of JAK2, whereby, MTIC exhibited the highest binding energy of − 42.885 kcal/mole, while the remaining ligands—TMZ, AIC and FU exerted a binding energy of − 32.530, − 33.159 and − 29.593 kcal/mole, respectively (Table 4).

Likewise, the docking analysis for STAT3 revealed that, MTIC depicted the highest binding energy of − 28.236 kcal/mole, while TMZ, AIC and FU exhibited binding energy of, − 20.667, 17.346 and − 23.658 kcal/mole, respectively (Fig. 7, Table 4). The binding energy of MTIC could be due to the formation of π-cationic interaction along with H-bond that contributed to the overall stability, when compared with the other anti-tumor ligands (Fig. 7).

Fig. 7.

Interaction profile of A STAT3, B IFN-γ and C NOS2 with ligands TMZ, MTIC, AIC and FU

In the case of IFN-γ, docking results revealed that AIC depicted the maximum interaction profile with a binding energy of − 36.282 kcal/mole (Fig. 7), while TMZ, MTIC and FU exerted a binding energy of − 31.600, − 24.403 and − 23.932 kcal/mole, respectively (Fig. 7, and Table 4).

With the most crucial target, NOS2, docking analysis revealed that AIC exhibited a highest binding energy of − 33.902 kcal/mole (Fig. 7), following, FU, TMZ and MTIC of − 24.978, -24.211 and − 20.902 kcal/mole (Fig. 7). This accounts for the presence of maximal number of H-bonds and salt bridge formation, as compared to other ligands and hence the maximum stability (Table 4).

To summarize, the complexity of bonding between the ligand–target protein complexes denotes the strength of molecular docking. From our visual analysis, based on the binding energy and docking score, it was evident that TMZ and its metabolites, MTIC and AIC were the best ligands to form complexes with TLR4, JAK2, STAT3, IFN-γ and NOS2, while FU exhibited highest interaction with IL-6. However, further investigations were warranted for verification of the stability of these ligand–protein complexes obtained by docking analysis.

Molecular dynamics simulation (MDS)

Following identification of the docked complexes and affirmation of their interaction profile, we then determined the stability of these interactions using molecular dynamics simulation (MDS). MDS was evaluated in terms of RMSD, RMSF, Rg and SASA against each of the docking complexes.

Among the MDS parameters, RMSD is used to measure the average change in displacement of atoms of a particular frame with respect to a reference frame. In general, the target reaches stability in few pico-seconds (ps). Initially RMSD was performed for IL-6, whereby we found that TMZ, MTIC and AIC depicted a stable interaction (0.5–0.59 nm) with the target protein till 20,000 ps, while FU exhibited a slight increase in the RMSD value (0.66 nm) from 15,000 ps (Supplementary Fig. 3). Similar to their interaction with IL-6, the ligands TMZ, MTIC and AIC showed a rise from (0.6–0.9 nm), following consistent RMSD plot (0.9 nm) for TLR4 as well, while FU portrayed a mild increase in RMSD value (1.2 nm) from 10,000 ps. Interestingly, all of the four ligands displayed an overall stability for JAK2 and STAT3 at 20,000 ps (0.4 nm). On the other hand, RMSD plots of IFN-γ and NOS2 revealed a stable RMSD value (0.2 nm) for all the chosen ligands, except AIC, which showed an elevation in the RMSD plot (IFN-γ—0.8 nm at 15,000 ps; NOS2—0.6 nm at 10,000 ps) (Supplementary Fig. 3). Based on the RMSD studies, we concluded that all of the chosen protein–ligand complexes had attained structural stability against all of these docked complexes.

To further verify the ligand–protein stability, we then evaluated the RMSF value for the ligand–protein interactions. RMSF plot enables to predict if the ligand–target interaction is sufficiently stable and provides basis for the flexibility of the conformation. RMSF fluctuation is often connected to RMSD plot, wherein more RMSF fluctuation yields less stable RMSD plot. Interestingly, we found that among the ligands that interacted with IL-6, TMZ, MTIC and AIC exhibited the minimum fluctuation (< 0.4 nm) while FU depicted the maximum fluctuation (0.7 nm) (Supplementary Fig. 3). In the case of TLR4, RMSF plot revealed that among all of the four ligands interacted, TMZ, AIC and FU showed the minimum fluctuation (< 0.8 nm) while MTIC exhibited maximum fluctuation of 1.0 nm. Conversely, the RMSF plots of JAK2 revealed minimal fluctuation in TMZ (0.4 nm) and maximum fluctuation for MTIC, AIC and FU (0.8 nm). On the other hand, amongst all of the docked complexes, MTIC, AIC and FU depicted a minimal fluctuation (< 0.4 nm) with IFN-γ, while TMZ demonstrated a slight increase in RMSF (at 200 residues). With regard to their interaction with NOS2, TMZ, MTIC and FU exhibited a minimum fluctuation of < 0.5 nm, while AIC showed the maximum fluctuation (2.5 nm). Taken together, in consensus with our RMSD results, the RMSF affirmed that the ligand–protein complexes were sufficiently stable and had low flexible conformation (Supplementary Fig. 3).

SASA is used to determine the surface area on a biomolecule that is accessible to solvent molecules. SASA plot denotes the ligand–protein interaction in the solvent. In general, low SASA value denotes that the target protein is less available for interaction with the solvent molecules like water; high SASA value on the other hand, implies that the target protein is widely accessible by the solvent molecules. Interestingly, the SASA plots of the interactions between all the four ligands and IL-6 revealed a low SASA value (100–140 nm²). Likewise, interaction between the ligands and TLR4 depicted a SASA value of 240–299 nm². On the other hand, interaction of the anti-tumor ligands with JAK2, STAT3 and NOS2 exhibited a SASA value of 285–305 nm², 285–305 nm² and 205–225 nm², respectively. In the case of IFN-γ, MTIC, AIC and FU exhibited a low SASA value of 100 nm², while TMZ exhibited slightly higher SASA value of 220 nm. Taken together, our SASA analysis revealed that most of the chosen proteins depicted a low SASA score and had no folding effect on complex stability (Supplementary Fig. 3).

Following elucidation of complex stability and surface exposure, the intrinsic dynamics of ligand–protein complexes were calculated by performing Rg on the docked complexes. It is well known that an increase in Rg values implies a decrease in the protein compactness leading to increased flexibility and less stability. Conversely, low Rg values on the other hand, denote stability and protein compactness. The Rg plots of all the chosen ligand–protein complexes are depicted in Supplementary Fig. 3. It can be seen that the Rg value of all the chosen ligands with IL-6 was measured between 1.55 and 1.70 nm with less fluctuation. In the case of TLR4, TMZ, MTIC and AIC depicted moderate high fluctuation and high Rg value between 3 and 3.4 nm at 10,000 ps, while FU depicted high fluctuation of 3.5 nm at 10,000 ps. For JAK2, MTIC and AIC depicted low fluctuation with Rg value ranging from 2.7 to 2.74 nm, while TMZ and FU demonstrated a high fluctuation rate of 2.76–2.8 nm. However, all of these four ligands depicted a high fluctuation, while Rg value for STAT3 ranging from 3.42 to 3.54 nm. For IFN-γ, TMZ depicted the highest Rg value of 2.6 nm, while MTIC, AIC and FU showed Rg value below 2 nm with low fluctuation.

Parallely, for NOS2, all the four ligands exhibited a low fluctuation and Rg value from 2.3 to 2.36 nm. Thus, from our Rg scores, we concluded that the chosen ligand–protein complexes had depicted sufficient stability and compactness.

Overall, our in silico data strongly established the fact the chosen anti-tumor ligands had formed stable and compact docking complexes with the inflammatory proteins and may be probed further, to modulate the inflammatory pathway in experimental setup as an affirmation of our in vitro findings.

Discussion

Glioma, accounting for about 40% of the intra-cranial tumours, is often associated with a poor prognosis and dismal survival rate. The current cutting-edge treatment modality encompasses surgical resection of the tumor mass trailed by radio and chemotherapy employing TMZ, the gold-standard drug for glioma (Oronsky et al. 2021). Though TMZ, as a mono-therapeutic agent had hailed a major breakthrough, its therapeutic efficacy has decreased over-years due to chemo-resistance. This highlights the critical necessity for efficient therapeutic approaches. Moreover, a number of preclinical studies had demonstrated that using mono-therapeutic drugs to target glioma was ineffective and was not linked to a significant improvement in patient outcomes (Ghosh et al. 2018). Combination therapy, which can produce a synergistic growth inhibitory impact while focusing on many gene targets at once, has thus emerged as the cornerstone of the current situation. For instance, a combination of TMZ with thymoquinone, a natural product had showed synergistic inhibition of proliferation of glioma cells, by inducing apoptosis (Pazhouhi et al. 2016). Pandey et al., reported that a cyclin-dependent kinase inhibitor, roscovitine, was known to potentiate the cytotoxicity of TMZ and reduced progression rate of glioma, either alone or in combination both in vitro and in vivo (Pandey et al. 2019). Above all, it has been put forward that, since TMZ is associated with considerable cytotoxic effects, synergizing TMZ with natural/synthetic derivatives would bring about concurrent reduction in the dosage and would be a beneficial approach for glioma treatment.

Derivatives of MNPs that are less toxic or non-toxic are frequently sought after in the quest for effective therapeutic candidates for combination therapy in glioma since they are linked to a low risk of side effects. One such MNP, FU, has been investigated as a possible anti-tumor agent in a variety of cancer cells. A recent study by Park et al., reported the plausible anti-inflammatory effects of FU through inhibition of NF-ƘB, MAPK and Akt activation in lipo-polysaccharide induced in BV2 microglial cells (Park et al. 2011).

Inflammatory cytokines, are one of the major attributes in the pathogenesis of glioma, that contributes to tumor proliferation and sustenance, thereby promoting cell growth in glioma. Interleukin-6 (IL-6), a pro-inflammatory autocrine and paracrine cytokine possess tumor promoting and progressing effects in glioma (Lanza et al. 2021). It acts directly on tumor cells by binding to interleukin-6 receptor (IL-6R) or glycoprotein complex 130 (gp130 complex) and activates Janus Kinases (JAK)—STAT3. STAT3, which belongs to a family of transcriptional factors, is involved in proliferation, inflammation and tumorigenesis in most of the cancer types, including GBM. On the other hand, inflammation mediated by binding of toll-like receptor (TLR-4), can induce expression of nitric oxide synthase isoform 2 (NOS2) which leads to the production of nitric oxide. Owing to the importance of IL-6 mediated JAK2/STAT3 signalling pathway in tumorigenesis, therapeutic targeting of this pathway has elevated the role of stratified treatment approaches in most of the cancers, including glioma.

Previously, TMZ and FU were reported to exert potent anti-proliferative and growth inhibitory effects, individually, in various cancer cells. In accordance with those findings, in the present study, the dual-drug combination (TMZ + FU) was found to be non-toxic in normal cells, however, when this cocktail was administered on glioma cells, the drugs significantly hampered the proliferation of glioma than their individual counterparts (Figs. 1 and 2). This may be attributed to the combination of the anti-proliferative effect of TMZ and growth inhibitory potency of FU. Further, to determine if such effective anti-proliferative effect was synergistic, we calculated the combination index for the dual-drug combination. We found that on average, 172.105-fold reduction of TMZ and 1.18-fold reduction of FU were sufficient to induce the median effect, while administering the drugs as a dual-drug combination (Fig. 2). Further, the SI index of the dual-drug combination was higher than that of the individual medications, indicating that the dual-drug combination specifically reduced the proliferation of glioma cells (Table 3). Glioma prognosis would be improved and patient survival would be maintained by achieving this synergistic anti-proliferative action at such a low medication dosage.

Proliferation and migration in glial tumours are mutually exclusive, which is an interesting observation. Alterations in the tumor micro-environment prompt glioma cells to “go/migrate” and re-settle to “grow/proliferate” in an adaptable environment (Hara et al. 2019). This “Go and grow” effect, which encourages tumorigenicity, is crucial in gliomas. Hence, it is conceivable that finding chemotherapeutic drugs that block glioma’s “Go” might aid in inhibiting tumour cell migration and subsequently reduce glioma proliferation (Zhou et al. 2022). According to previous studies, TMZ alone prevented glioblastoma from migrating in vitro in U251 (Du et al. 2018). Moreover, HepG2 liver cancer cell migration and proliferation were hindered by FU. In accordance with these findings, our results revealed that both the drugs had potent anti-migratory effect against glioma (Fig. 3, Supplementary Fig. 4). Interestingly, using these treatments in combination had significantly preserved the cell-free space, stopping the “Go” phenomenon and reducing glioma proliferation (Fig. 3). The combination of TMZ + FU may have at least partially suppressed the matrix metalloproteinases (MMPs), which is one potential mechanism for the anti-migratory impact demonstrated by this dual-drug treatment. As an interesting note, the dual-drug combination of TMZ and FU regulated the expression of MMPs in BALB/C mice model (U87MG-fLuc transfected model) (Li et al. 2018) and HT1080 fibro sarcoma cells (Teruya et al. 2019). However, profound molecular analysis is required to conclusively determine if the anti-migratory effect of this combinatorial treatment was due to the suppression of MMPs.

The anti-proliferative and the anti-migratory effects of the combinatorial drug treatment might be due to induction of apoptosis. In this perception, we perceived that our dual-drug combination (TMZ + FU) exhibited maximum alteration at nuclear and cytoplasmic level (Supplementary Figs. 1 and 2), which can be triggered by activation of apoptotic signalling pathways, in comparison to control and individual-drug treated groups. Most cancer cells including GBM exhibits evasion of apoptosis and transforms to a high grade of malignancy that generates therapy-resistance (Mohammad et al. 2015). Therefore, restoration of apoptotic pathway by therapeutic targets is one of the promising challenges faced in the field of anti-glioma drug development, while their exact mechanism of action has not been clearly demonstrated. The key mechanism that cascades of caspase-mediated events propagate cell death through apoptosis is widely understood. Caspases are prevalent chemotherapeutic or chemo preventive targets, from the perception of anti-cancer therapeutics (Pfeffer and Singh 2018). Most of the anti-cancer agents derived from marine resources have shown to induce apoptosis by activation of caspase-3 (Pfeffer and Singh 2018).

According to previous findings, TMZ may cause apoptosis through the non-canonical dependent mechanism of STAT3 mitochondrial translocation (Cui et al. 2020). In a similar study, it was reported that STAT3 suppression leads to enhanced sensitization of glioma by TMZ via down regulation of MGMT expression (Kitange et al. 2009).

On the other hand, both JAK/STAT3 signalling and autophagy promotion by FU were known to cause apoptosis (Zhang et al. 2020a, b). In addition, FU was shown to induce apoptosis in liver cancer cell line HepG2 by down regulation of p-STAT3 (Roshan et al. 2014). In accordance with these findings, we observed that dual-drug combination of TMZ and FU led to enhancement of caspase-3 expression significantly, in comparison to individual drug treated and control groups. Thereof, we postulate that the combinatorial treatment of TMZ and FU augmented apoptosis by elevating the levels of caspase-3, possibly by down regulating the expression of JAK/STAT3 (Fig. 5).

Further, to elucidate the possible binding mechanism of each of these ligands, TMZ, FU, MTIC and AIC, with the pro-inflammatory targets—IL-6, TLR4, JAK2, STAT3, IFN-γ and NOS2, individually, molecular docking was performed. Docking performed of TMZ, MTIC, AIC and FU at the active sites of the target proteins were analysed using parameters, such as Glide based binding energy, hydrogen bond, π-cation, and salt bridge interactions. The binding energy revealed that FU-IL-6 exhibited highest binding energy, as compared to the other ligand–IL-6 complexes. Each of the complexes (IL-6-TMZ, MTIC, and AIC) was stable during the 20,000-ps run time period, as per simulation studies, which could have been generated by the creation of H-bonds (Supplementary Fig. 3, Fig. 6).

Furthermore, we observed that (TLR4, JAK2 and STAT3)–MTIC; (IFN-γ and NOS2)–AIC binds strongly, as evident by their highest binding energy values. We found major л-cations interactions among STAT3–MTIC and AIC; NOS2–TMZ and MTIC complexes. Additionally, there were formation of salt bridges among (TLR4–TMZ, MTIC and AIC; JAK2–TMZ and AIC; STAT3–TMZ and AIC; IFN-γ-AIC; NOS2–AIC) except IL-6, which suggests that apart from H-bond, salt bridges and л-cation interactions, may contribute to overall stability of these complexes. Studies have reported that salt bridges and л-cation interactions are non-covalent form of interactions, which might exhibit more strong interactions than H-bond (Fink and Boratyński 2014). Interestingly, TMZ along with MTIC and AIC participated in salt bridge and л-cation interactions, in most of the targets, with an exception to FU. This could be attributed to the presence of L-fucose and sulphate ester groups in FU, which might attribute to its cytotoxic and anti-cancer activity (Wang et al. 2020), Since TMZ and its metabolites have imidazole and triazene rings, they can develop other non-covalent interactions that induce conformational changes in the targets that lead to inflammation. However, the exact mechanism of induction of apoptosis is still yet to be studied in detail in in vivo experimental models as well.

Nevertheless, there are few limitations in the current study. Firstly, the glioma cell line employed in this study is not resistant to TMZ. Thereof, expansion of this work in TMZ resistant cell lines is warranted. Second, effectiveness of the proposed combinatorial treatment in an in vivo setup is prerequisite to strengthen our in vitro and in silico findings. In addition, the effect of these drugs on the anti-apoptotic markers and other pro-apoptotic signalling cascades are to be explored in mere future.

Conclusion

The key to accomplish successful drug repurposing is a strategy to discover the most effective drug. Here, we have demonstrated the efficacy of a marine derivative, FU on enhancing the anti-glioma potency of TMZ both in an in vitro and in silico setup. Our results, on a nutshell, indicated that the combinatorial treatment of TMZ + FU had synergistically inhibited glioma proliferation via modulation of inflammatory pathway and paving way for apoptosis. Hence, we speculate that FU could serve as a potential candidate for conventional glioma treatment enhancement.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

ATS: conceptualization, resources, methodology, supervision, writing—review and editing. IB: data curation, experimentation, writing-original draft preparation. DSP & SSK: data curation. MAR: data visualization, SA: data curation, VR: data curation, DP: data visualization. All authors reviewed the manuscript.

Funding

The authors declare that no funds, grants or other support were received during the preparation of the manuscript.

Data availability

The data that support the findings of the present study will be available upon requesting corresponding author (T.S Anitha, through mail) under specific circumstances.

Declarations

Conflict of interest

No conflict of interest is reported by authors.

Compliance with ethical standards

None.