Introduction

Microtubules, the dynamic polymers of eukaryotic cytoskeleton, play important roles in several cellular functions including cell division, cell differentiation, cell migration, intracellular trafficking and cell signalling cascades1,2,3,4. Microtubules are polymers of tubulin heterodimers (αβ tubulin), and conformational transitions in the microtubule lattice drive microtubule dynamic instability and affect various aspects of microtubule function5,6,7. Several clinically used anticancer drugs target microtubules network and block the cell cycle progression in mitosis, which eventually activates the cell death program8,9. These anti-microtubule based cancer drugs modulate the dynamics of microtubules either by depolymerizing or by increasing the polymerization rate of the microtubules8,10,11. Based on the mechanism of action these drugs are broadly classified into two groups; (i) microtubule stabilizing drugs such as taxol, epothilone or docetaxel and (ii) microtubule depolymerizing drugs such as vinblastine, vincristine or colchicine12,13. Tubulin heterodimer has distinct binding site for these drugs. There are three well characterize sites are present on tubulin heterodimers; (i) taxanebinding site; present on the luminal site of β-tubulin, (ii) vinca binding site; present on the tip of β-tubulin and (iii) colchicine binding site; present at the interface of α-β tubulin heterodimers14,15,16. The exact nature of these transitions and their modulation by anti-cancer drugs are still a matter of research because capturing the real-time dynamics at high resolution is very challenging17,18,19. Though microtubule targeting drugs are found to be very successful in cancer chemotherapy, the development of resistance against the existing drugs limits the use of known tubulin-targeting drugs in cancer chemotherapy8,20,21. So, the quest for discovering new agents having chemotherapeutic potential has always been a part of cancer chemotherapy research.

Hydroxyethylamine(HEA) is a well-known scaffold in medicinal chemistry that is of significant importance22,23. It serves as a key structural element in the design of various pharmaceutical compounds due to its ability to mimic peptide bonds and interact with enzymes and receptors in specific ways24,25. Some notable applications of HEA-based analogues in medicinal chemistry include HIV-1 protease inhibitors, antimalarial agents, antiviral agents, β-Blocker activity, etc26,27,28.The versatility of HEA-based analogues lies in their adaptability to different pharmacological targets by modifying their chemical structure while preserving the essential HEA moiety24. Medicinal chemists can fine-tune these compounds to optimize their interactions with specific biological targets, making them valuable tools in drug discovery and development for various therapeutic areas. Herein, we screened the library of 106 biologically active compounds in silico to identify potent inhibitors which target microtubule assembly dynamics. The hydroxyethylamine (HEA) analogues are the best hit among all the screened analogues. We modified the HEA inhibitors to find more potent inhibitors which can potentially target the microtubules. Similar to colchicine, the modified compoundBKS3031A(2S, 2’S,3R,3’R)−1,1’-(piperazine-1,4-diyl)bis(3-amino-4-phenylbutan-2-ol), BKS3045A(S)−2-amino-N-((2R,3 S)−4-(4-(4-bromobenzyl)piperazin-1-yl)−3-hydroxy-1-phenylbutan-2-yl)−3-phenylpropanamide), and BKS3046A(S)−2-amino-N-((2R,3 S)−4-(4-(4-bromobenzyl)piperazin-1-yl)−3-hydroxy-1-phenylbutan-2-yl)−3-methylbutanamide) were found to bind at the binding interface of α,β tubulin heterodimers. Thereby, it may inhibit the microtubule assembly dynamics.

Materials and methods

  1. (1)

    Molecular docking experiment.

Molecular docking study was performed using the Schrodinger Glide program. A library of 106 biologically active scaffolds, including chalcones, small amides and hydroxyethylamines (HEA), was compiledfrom the literature29,30,31,32. These compounds were prepared for protonation at biological pH. Further, we have performed isomerisation, conformation minimization using the Ligprep module. PDB ID: 1SA0 is retrieved from the protein data bank and scrutinized for the missing residues. The reference compound colchicine binds at the interface region of ɑ/β tubulin complex, thus, chains A and B chains were preserved and the remaining were removed from the complex structure. The ɑ/β tubulin complex was further prepared using the Protein Preparation Wizard and colchicine binding region residues were defined to generate the grid box generation. Three steps (HTVS, SP and XP) processes of the glide program were used to refine the hit compounds and top 10 hits from XP were used for visual interaction analysis.To enhance the affinity of hit compounds towards the receptor, we further did deprotection of phthalimide and tertiary butyl carbonate. The IUPAC name and structure of all the inhibitors are listed in supplementary table S1.

(2) Molecular dynamics Simulation.

Molecular Dynamics (MD) simulation is utilized to explore the dynamics of the protein in complex and receptor only forms. The original protein structure was obtained from the Protein Data Bank (PDB-ID: 1SA0), and only A and B chains were preserved. Simulations were run using the Amber 16 software package, employing the ff14SB force field for protein parameterization33. Parameters for the docked molecules were generated by applying the antechamber using the GAFF force field. The protein was positioned at the center of a 10Å cubic box filled with TIP3P water followed by adding the appropriate number of Na+ and Cl- ions replacing some water molecules to neutralize the system.

During the equilibration, each simulation box first underwent energy minimization for water, neutralizing ions, followed by minimization of the protein complex. The system was then gradually heated from 50 K to 300 K in 50 K increments. After heating, the system was equilibrated using the NVT ensemble for 100 ps, followed by a 200 ns production run under the NPT ensemble. The SHAKE algorithm was applied to constrain hydrogen atoms, while long-range electrostatic interactions were handled using the Particle Mesh Ewald (PME) method. Temperature control was managed with Langevin dynamics, and the pressure was regulated using the Berendsen barostat.

(i) MD trajectories analyses.

The obtained MD trajectories were analysed for the structural stability of ɑ/β tubulinand binding with drug molecules through root mean square deviation (RMSD), root mean square fluctuation (RMSF), radius of gyration (Rg), solvent-accessible surface area (SASA) and H-bond occupancy between ɑβ-tubulin and BKS3031A, BKS3045A, and BKS3046A during the simulation. The free binding energy estimations of BKS3031A, BKS3045A, and BKS3046A were calculated using the generalized Born model and solvent accessibility method (MM/GBSA) along with the weighted interactions active-site residues34,35,36,37,38.

(ii)Calculation of binding interactions using MM-GBSA.

The binding free energies of BKS3031A, BKS3045A, and BKS3046A to ɑβ-tubulin were calculated using the generalized Born model and solvent accessibility method (MM-GBSA) along with the weighted interactions active-site residues34,35,36.The binding free energy components can be represented according to the Eq. 

$$\:\varDelta\:{\text{G}}_{\text{b}\text{i}\text{n}\text{d}}={\text{G}}_{\text{c}\text{o}\text{m}\text{p}\text{l}\text{e}\text{x}}-\left({\text{G}}_{\text{r}\text{e}\text{c}\text{e}\text{p}\text{t}\text{o}\text{r}}+{\text{G}}_{\text{l}\text{i}\text{g}\text{a}\text{n}\text{d}}\right)$$
(1)
$$\:\varDelta\:{\text{G}}_{\text{b}\text{i}\text{n}\text{d}}=\:\varDelta\:\text{H}-\text{T}\varDelta\:\text{S}\:\sim\:\varDelta\:{\text{E}}_{\text{M}\text{M}}+\varDelta\:{\text{G}}_{\text{s}\text{o}\text{l}}-\text{T}\varDelta\:\text{S}\:$$
(2)
$$\:\varDelta\:{\text{E}}_{\text{M}\text{M}}=\:\varDelta\:{\text{E}}_{\text{i}\text{n}\text{t}\text{e}\text{r}}+\:\varDelta\:{\text{E}}_{\text{e}\text{l}\text{e}\text{c}\text{t}\text{o}\text{s}\text{t}\text{a}\text{t}\text{i}\text{c}}+\varDelta\:{\text{E}}_{\text{v}\text{d}\text{w}\:}$$
(3)
$$\:\varDelta\:{\text{E}}_{\text{s}\text{o}\text{l}}=\:\varDelta\:{\text{G}}_{\text{G}\text{B}}+\:\varDelta\:{\text{G}}_{\text{S}\text{A}}\:$$
(4)

Where ∆EMM represents the enthalpic components, while ∆Esol represents the polar and non-polar electrostatic components from solvation. Here, polar electrostatic component is calculated using the GB model, while the non-polar electrostatic contribution is calculated by solvent accessible surface area (SASA). The last 50 ns simulation trajectory is used, which was sampled per 10 ps interval.

Results & discussion

With the advancement in recent computational methods, virtual screening approach has been widely employed as a starting point for identifying the suitable drug candidate molecules. Designing and development of new generation drug molecules as a potential inhibitor against the target proteins associated with diseases39,40,41,42,43. Overall, virtual screening serves as a valuable bridge between computational methods and experimental drug discovery. While it provides a cost-effective and rapid means of identifying potential drug candidates, it is typically followed by rigorous experimental validation and optimization to ensure the safety and efficacy of the identified compounds for the treatment of diseases associated with specific target proteins40,43,44,45. HEA derivatives containing aromatic sulphonamideshave already known for their cytotoxic potential against several cancerous cell lines. Due to their potential ability against cancer cell proliferation, it has gained attention to identify its analogue to find out more potent anticancer agents. In this study, to identify the promising candidates against αβ-tubulin, virtual screening of 106compounds were performed using GLIDE module of Schrodinger which provides a rational three tier filtering of molecules.Virtual screening of compounds against αβ-tubulin colchicine binding pocket results in the selection of six HEA compounds as plausible tubulin inhibitors, having the molecular binding scores >−10.0 kcal/mol. The best hit compounds BKS3046 ((1R,2 S)-N-((2R,3R)−4-(4-(((1s,4 S)−4-bromocyclohexyl)methyl)cyclohexyl)−1-cyclohexyl-3-hydroxybutan-2-yl)−2-((1R,3R,3aR,7aS)−1,3-dihydroxyhexahydro-1 H-isoindol-2(3 H)-yl)−1-hydroxy-3-methylbutan-1-aminium) with docked score= −11.45kcal/mol is spatially binds to tubulin at the colchicine binding site, stabilizing the ɑ/β-heterodimer46,47. In αβ-tubulin, the colchicine binding site is located at interface of α/β-subunits opposite to the GTP binding pocket which is observed consistent in the docking studies (Fig. 1). The hydroxy groups of BKS3031 substituted with iso-indole is stabilized by polar molecular interaction with residues, GLN11, TYR224 and GLN245. The bromocyclohexene and cyclohexylbutan moiety is deeply fitted in hydrophobic grooves of active site, surrounded by VAL177, ALA180, LEU246, LEU253, MET257, ALA314, ALA315, VAL316, ALA352, whereas protonated ammonium (–NH2+–) is engaged in cation- polar interaction with GLN245 (Fig. 2).The molecular interactions of remaining top five compounds are shown in Supplementary Figure S1.

Fig. 1
Fig. 1
Full size image

Cartoon view of molecular binding of ligands with ɑβ-tubulin. The molecular binding of Colchicine, GTP, BKS3031 and BKS3031A, defining the molecular interaction of BKS3031 and BKS3031A at the colchicine binding site located at the interface of ɑ and β subunits of tubulin.

Fig. 2
Fig. 2
Full size image

Molecular interaction of BKS modified compounds at the colchicine binding pocket. (A) colchicine (B) BKS3031A (C) BKS3045A (D) BKS3046A.

These top ranked six compounds were further modified to enhance their activity as enumerated in Table 1. All the top six hit compounds possess the phthalimide group, which is bulkier, hydrophobic, and less polar. The phthalimide group can be transformed into a primary amine functional group by using hydrazine hydrate according to the literature procedure (https://doi.org/10.1002/0471220574.ch7). The change in functional groups facilitated the better occupancy of compounds at the binding pocket. Results show an increment of ~ 1.0–2.0 kcal/mol in the binding affinity of compounds, except BKS3046 (−11.45 kcal/mol) which shows marginal decrease in binding affinity (−11.14 kcal/mol). Of the selected six compounds, three compounds (BKS3031A, BKS3045A, BKS3046A) were ranked top. However, the compound BKS3045 (−10.64 kcal/mol) shows a remarkable change in binding affinity after modification with the highest docking score of −12.14 kcal/mol (Fig. 2). The substitution of dihydroxyisoindolin with N-methyl-1-phenylmethanamin shows a better yield in the binding of BKS3045A as compared to BKS3031A (−11.25 kcal/mol) and BKS3046A (−11.18 kcal/mol).

Table 1 Molecular docking scores of compounds at the colchicine binding sites of ɑβ- tubulin.
Full size table

Although the GLIDE is a very popular molecular docking program, uses hybrid approach, protein flexibility model, exhaustive ligand flexibilities and a series of hierarchical filtering strategy for the selection of ligand binding poses and spatial orientations. However, it may fail to sample the global flexibility of protein and loops, and loops containing active site side chains mobilities47,48. In the context, MD simulation approach allows to evaluate residues flexibility at the atomic resolution. It can efficiently handle the flexibility of both the ligand and protein, capture the receptor conformations, receptor-ligand atomic flexibility, structural dynamics and stabilities in the solvent aqueous environment. Moreover, using the MD trajectories of protein-ligand complexes, a consensus scoring can be defined in terms of binding free energies which is more accurate for the prediction of bound conformations and poses of protein-ligand interactions and binding affinity34,49. Thus, to examine the molecular binding stability of protein-ligand interactions, all atoms MD-simulations were performed on the docked complexes of the best three compounds for the time period of 200 ns. Furthermore, the obtained trajectories were subjected to estimate the free binding energy of compounds with ɑβ tubulin.

MD simulation

Molecular dynamics (MD) simulation is a widely accepted and reliable computational technique used to study the behaviour and interactions of atoms and molecules over time. It can provide atomic-level insights of the molecular interactions. To understand the structural dynamics and stability of BKS3031A, BKS3045A, and BKS3046Aat colchicine binding sites of ɑβ- tubulin, all atoms MD simulations were performed in aqueous environment for the period of 200 ns at the physiological temperature, 300 K. The structural dynamics and stability of ɑβ-tubulin heterodimer complexed with BKS3031A, BKS3045A, and BKS3046A were determine by RMSD, Rg, RMSF and SASA. The average values of structural parameters are enumerated in Table 2. The molecular binding stability of compounds analysed by time evolution plots of H-bonds interactions. In addition to structural studies, we also performed MM-GBSA to estimate the binding free energy of BKS3031A, BKS3045A, and BKS3046A. Furthermore, the decomposition plot is used to illustrate the contribution of active site residues, stabilizing the molecular interaction of BKS3031A, BKS3045A, and BKS3046A compounds with ɑβ-tubulin in terms of binding free energy (∆G).

Table 2 Average structural order parameters values of αβ-tubulin.
Full size table

In order to examine the structural stability of protein, the backbone Cα-atoms RMSD were calculated for ɑβ-tubulin complexed with colchicine and BKS3031A, BKS3045A, and BKS3046A. The average change in distance between the backbone atoms during the simulation assessing stability the of system which is inversely correlated.The less variation in RMSD pronounced a stable structural dynamic of protein-ligand complex39,50. Figure 3 shows that the complexes of ɑβ-tubulin with BKS3031A, BKS3045A, and BKS3046A achieve equilibrium very quickly and remain stable till the simulation finished at 200 ns, ranging the average RMSD value approximately ~ 1.87–1.97Å. Relatively the colchicine docked structure of ɑβ-tubulin optimized around ~ 35 ns, the RMSD trajectory observed stable up to ~ 150 ns. A slight dropdown in RMSD can be seen at ~ 150 ns, however the trajectory acquired equilibrium and remains stable up to 200 ns. The docked structure of ɑβ-tubulin with colchicine shows average RMSD value 3.01 ± 0.29 nm whereas the structures with BKS3031A, BKS3045A and BKS3046A attain equilibrium around the average RMSD values 1.87 ± 0.15, 1.97 ± 0.14 and 1.80 ± 0.12 nm, respectively. The stable RMSD trajectory of all four complexes indicate that ligands were well accommodated at the active site of ɑβ-tubulinand protein has not undergone substantial alterations, during the period of MD simulation.

Fig. 3
Fig. 3
Full size image

Time evolution plots of backbone-backbone RMSD in water at 300 K for ɑβ-tubulin and its complexes with ligands BKS3031A, BKS3045A, BKS3046A and Colchicine. The color codes for the ɑβ-tubulin docked complexes with ligands are given in figure inset at top.

To access the dynamic stability of ɑβ-tubulin docked complexes we also measure the average distance compounds from the centre of active site. Results show the average distance of all compounds varying between the 3.85 − 5.20 Å which signify that the compounds were well accommodated at the active site of ɑβ-tubulin (Fig. 4). The binding site of colchicine is at the ɑ/β interface this may be the reason colchicine shows higher average distance 5.20 ± 0.15 Å. The conformational dynamics of BKS3031A is observed around the lowest average distance of ~ 3.85 ± 0.09 Å. Whereas, the average distance of BKS3045Aand BKS3046Ais observed restricted around 4.82 ± 0.21 Å and 4.13 ± 0.11 Å, respectively. Thus, all four compounds remain spatially occupied at the active of ɑβ-tubulin which suggest the stable protein-ligand binding interaction during the simulation. Furthermore, the higher average distance of colchicine due to the interfacing binding also exemplify the reason to achieve delayed RMSD equilibrium by ɑβ-tubulin– colchicine complex and the relatively agitated RMSD trajectory as compared to ɑβ-tubulin complexed with BKS3031A, BKS3045A, and BKS3046A.

Fig. 4
Fig. 4
Full size image

Average distance of all compounds varying between the 3.85 − 5.20 Å which signify that the compounds were well accommodated at the active site of ɑβ-tubulin. The color codes representing the docked complexes with ɑβ-tubulin are given in figure inset at top.

To measure the dynamic stabilities of ɑβ-tubulin docked complexes we also calculated the radius of gyration (Rg) which determine the compactness changes of a ligand-protein complex. Results shown in Fig. 5 clearly decipher the stable Rg trajectory of all four ɑβ-tubulin docked complexes. The Rg-plot of ɑβ-tubulin– colchicine attain equilibrium during the initial ~ 20 ns and remains stable throughout the simulation period around the average Rgvalue 21.76 ± 0.10Å. The structure of ɑβ-tubulin docked with BKS3031A, BKS3045A, and BKS3046Aachieve equilibrium quicklyat the beginning of the simulation and shows minimal agitation throughout the simulation period.BKS3031A docked complex observed stable with the Rg value 21.32 ± 0.05 Å whereas the docked structures with BKS3045A and BKS3046A are observed stable around the Rg values 21.51 ± 0.06 Å and 21.30 ± 0.05 Å, respectively.

Fig. 5
Fig. 5
Full size image

Time evolution plots of the radius of gyration (Rg) for ɑβ-tubulin and its complexes with ligands BKS3031A, BKS3045A, BKS3046A and Colchicine. The color codes for the ɑβ-tubulin docked complexes with ligands are given in figure inset.

Another structural parameter SASA is utilised to examine the solvent accessibility of docked complexes. SASA potentially attributable to interaction between complexes and solvents. In Fig. 6, it can be seen that due to the interfacial binding of colchicine, the ɑβ-tubulin- colchicine has delayed stabilization and showing the higher fluctuation. Whereas the burial binding of BKS3031A, BKS3045A, and BKS3046A is showing relatively more stabilized trajectory as compared to colchicine. The ligand binding site of ɑβ-tubulinis located at the ɑ/β domain interface region, providing considerable geometric space to accommodate ligands. Remarkably, the average distances of all four ligands remain consistent ≤ 5.0Å, suggesting the stable molecular interactions with ɑβ-tubulin. Defining the molecular binding stability of protein-ligand complexes, localized flexibility of individual amino acid residues throughout the residue sequence can be obtained from RMSF calculations.

Fig. 6
Fig. 6
Full size image

Time evolution plots of solvent accessible surface area (SASA) of ɑβ-tubulin and its complexes with ligands BKS3031A, BKS3045A, BKS3046Aand Colchicine. The color codes for the ɑβ-tubulin docked complexes with ligands are given in figure inset.

The RMSF analysis provides a clue about the extent of average atomistic fluctuations in individual residue of the entireprotein while the evolution of simulation. The high RMSF value indicates high degree of fluctuations which generally belongs to loop regions. In general, the residues at N-and C-terminals of the protein also show high RMSF values.Compared with loops, the regions belonging to stable conformations of ɑ-helices and β-sheets, and constrained regions having lower RMSF values. Figure 7 shows that along with the secondary structures of ɑ-helices and β-sheets, the active site residues also having the average atomic fluctuation ≤ 2.0 Å. The hydrophobic cavity comprising residues VAL177, ALA180, LEU246, LEU253, MET257, ALA314, ALA315, VAL316, ALA352 and the residues ASN101, SER178, THR179, TYR224, GLN247 observed participating in polar interactions showing the low atomic flexibility.

Fig. 7
Fig. 7
Full size image

RMSF plots for ɑβ-tubulin and its complexes with ligands BKS3031A, BKS3045A, BKS3046A and Colchicine. The color codes for the ɑβ-tubulin docked complexes with ligands are given in figure inset.

Further, we investigated the hydrogen bond (H-bond) interaction between the ɑβ-tubulin and compounds. For the molecular binding of ligands at active site H-bond interactions play a crucial role. The occupancy of H-bond interactions during the simulation is shown in Fig. 8 which is calculated with distance cutoff of 3.5 Å, and angle cutoff of 135◦. The molecular binding of colchicine shows the maximum occupancy of two bond interactions at the active site of ɑβ-tubulin, however, only one H-bond observed consistent during the simulation. The compound BKS3045A shows the maximum number of H-bonds interactions. During the initial 0–50 ns, it shows four H-bonds interactions. The number H-bond interactions increase up to six with the progression of simulation. however, with the intermittent appearing and disappearing of H-bonds, only three H-bonds remain stable during the simulation. The molecular binding of BKS3031A shows the maximum propensity of five H-bonds which can be seen during 0–50 ns, consequently reduced to three H-bonds during the 50–200 ns. BKS3045A also shows the occupancy of six H-bonds, of these only three H-bonds prolonged till the simulation finished at 200 ns. During the initial 0–50 ns, it shows the interactions of three H-bonds which is shifted to five at ~ 50–150 ns and infrequent appearing of seven H-bonds during the last ~ 50 ns of simulation. Thus, H-bonds analysis allows investigating the molecular interaction stability of compounds at the active site of ɑβ-tubulin.

Fig. 8
Fig. 8
Full size image

Time evolution plots of hydrogen bond interactions between ɑβ-tubulin and its complexes with ligands BKS3031A, BKS3045A, BKS3046A and Colchicine. The color codes for the ɑβ-tubulin docked complexes with ligands are given in figure inset.

MM-GBSA (best compound: BKS3031A)

To determine the molecular binding interaction and stability of BKS3031A, BKS3045A, and BKS3046A with ɑβ-tubulin, the binding free energy is computed using MM-GBSA. MM-GBSA provides the accurate estimation of binding energy in terms of different energy components in terms of polar and non-polar solvation free energy, bonded, non-bonded, electrostatic and van der Waals (vdW) interactions. According to the binding free energy BKS3031A (ΔGBIND=−32.67 kcal/mol) shows the stronger the interaction with ɑβ-tubulin as compared to colchicine (ΔGBIND =−25.86 kcal/mol). Result shows that the binding abilities of two inhibitors, BKS3045A (−21.77 kcal/mol)BKS3046A (−22.92 kcal/mol) having the weaker binding abilities to ɑβ-tubulin as compared to colchicine. The binding free energies of all four compounds with energy components are shown in Table 3. From Table 3, it can be seen that the energy components ΔEvdW and ΔEEEL provide favourable contribution to the bindings of compounds at the active site. Although with polar interactions, hydrophobic interactions are crucial for the binding stability of ligands at the active site of ɑβ-tubulin. The compound BKS3031A shows higher van der Waals (ΔEvdW=−50.53 kcal/mol) and the lowest ΔEvdWvalue − 39.83 kcal/molis obtained for BKS3045A. BKS3046A shows higher ΔEvdW(−45.53 kcal/mol) and ΔEEEL (−32.62 kcal/mol) as compared to Colchicine (ΔEvdW= −40.45 kcal/mol; ΔEEEL= −32.62 kcal/mol). However, the favourable binding energy of ΔEvdWand ΔEEELare completely screened by the stronger polar solvation energy, which results in the less favourable total binding energy of BKS3045A and BKS3046A as compared to Colchicine. It is also remarkable to note that in the molecular docking analyses BKS3045A shows higher binding affinity to ɑβ-tubulin (Table 1), however, applying the MD simulation which offers the broader flexibilities to protein and ligands andaqueous solvation to system allowed to estimate dynamic nature of the binding events which suggest the better affinity of BKS3031A.Finally, to measure the contribution of active site residues stabilizing the ligand, the energy decomposition plot defining the interaction energies was computed which evident the major role of hydrophobic residues in the binding stability of compounds (Fig. 9). Thus, the molecular docking followed by MD simulation and MM-GBSA provided a comprehensive overview of protein-ligand interactions and the role of water molecules in structure-baseddrug development processes34,49.

Table 3 Estimated binding free energy (kcal/mol) of docked complexes using MM-GBSA.
Full size table
Fig. 9
Fig. 9
Full size image

The binding free energy plot, representing the contributions of active site amino acids of ɑβ-tubulin involved in molecular interactions with the ligands, BKS3031A, BKS3045A, BKS3046A and Colchicine.

Conclusion

Our study aimed to identify HEA analogues as potential inhibitors targeting microtubule assembly dynamics. Through the modification of HEA inhibitors, specifically the deprotected BKS3031A, BKS3045A, and BKS3046A, we have successfully designed compounds that exhibit a higher affinity for microtubules. These modified compounds, have demonstrated their ability to bind at the binding interface of α-β tubulin heterodimers, suggesting their potential to inhibit microtubule assembly dynamics. Our findings shed light on the development of novel inhibitors with the potential to disrupt microtubule function, a critical process in cellular division and various cellular functions. The successful design and characterization of these analogues provide a foundation for further research into their therapeutic applications, such as in the treatment of cancer, where microtubule dynamics play a crucial role in cell proliferation. Moreover, this study highlights the significance of structure-based drug design in the development of targeted therapies. The utilization of molecular modelling and chemical modifications has enabled us to enhance the binding affinity of HEA analogues, opening exciting avenues for future drug discovery and development efforts.

In summary, our work represents a significant step towards the development of HEA analogues as potential microtubule-targeting inhibitors and underscores the importance of rational drug design in advancing therapeutic interventions in various diseases.”