Identification of Beclin-1 inhibitors to treat Immune thrombocytopenia by structure-based virtual screening and molecular modelling approaches

RuiBo Zhang; Juan Peng; Lan Ren; Jingjing Cheng; YiGang Guo; Kun Zheng

doi:10.25259/JKSUS_917_2025

View/Download PDF

Buy Reprints

PDF

Translate this page into:

Research Article

2025

:37;

9172025

doi:

10.25259/JKSUS_917_2025

Identification of Beclin-1 inhibitors to treat Immune thrombocytopenia by structure-based virtual screening and molecular modelling approaches

RuiBo Zhang^a,#,, Juan Peng^b,#, Lan Ren^a, Jingjing Cheng^c, YiGang Guo^d, Kun Zheng^a,*

aDepartment of Pediatrics, Taihe Hospital, Hubei University of Medicine, Shiyan, People’s Republic of China

bDepartment of Blood Transfusion, Taihe Hospital, Hubei University of Medicine, Shiyan, People’s Republic of China

cDepartment of Cardiac Function, Taihe Hospital, Hubei University of Medicine, Shiyan, People’s Republic of China

dDepartment of Hematology, Taihe Hospital, Hubei University of Medicine, Shiyan, People’s Republic of China

^#The two authors contribute equally

* Corresponding author: E-mail address: zhangruibo@taihehospital.com (K. Zheng)

Received: 2025-05-22, Accepted: 2025-10-08, Epub ahead of print: 2025-11-26, Published: 2025-11-29

Licence

This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-Share Alike 4.0 License, which allows others to remix, transform, and build upon the work non-commercially, as long as the author is credited and the new creations are licensed under the identical terms.

Abstract

Immune thrombocytopenia (ITP) is an autoimmune condition characterized by platelet loss, with autophagy-related proteins such as Beclin-1 significantly contributing to pathogenesis. This research employed a computational structure-based virtual screening (SBVS) approach to identify novel inhibitors that target Beclin-1 as prospective therapies for ITP. A comprehensive virtual screening of an Enamine-AI-Enabled-TF library, succeeded by molecular docking to evaluate binding affinities. ADMET profiling was performed to assess the pharmacokinetic properties (molecular weight (MW), lipophilicity (LogP), total polar surface area (TPSA), solubility (LogS), and potential toxicity risks, including mutagenicity, carcinogenicity, and reproductive toxicity), resulting in the identification of two promising candidates: Z9255356311 and Z9255355469. These candidates were selected based on their advantageous characteristics in absorption, distribution, metabolism, excretion, and toxicity. Molecular dynamics (MD) simulations confirmed the structural stability and binding persistence of these compounds within the Beclin-1 binding pocket. The results establish a robust foundation for subsequent in vitro and in vivo validation, facilitating the advancement of Beclin-1-targeted therapies for ITP.

Keywords

Autophagy-related proteins

Beclin-1

Immune thrombocytopenia (ITP)

Molecular docking

MD simulation

Virtual screening

Show Related Articles from PubMed

1. Introduction

Immune thrombocytopenia (ITP) is characterized as an autoimmune disorder that leads to a diminished platelet count. This condition arises due to an elevated rate of platelet destruction coupled with a reduction in platelet production (Li et al., 2015). The pathogenesis of ITP is complicated, characterized by the presence of autoantibodies directed against platelet surface glycoproteins, the cytotoxic T-cell-mediated lysis of platelets, and abnormalities within the bone marrow microenvironment (Kashiwagi and Tomiyama 2013; Zhou et al., 2015). Present therapeutic approaches encompass corticosteroids, intravenous immunoglobulin (IVIG), thrombopoietin receptor agonists, and immunosuppressive drugs. Nevertheless, these medications frequently exhibit restricted efficacy, considerable side effects, or require prolonged administration, underscoring the pressing demand for innovative, focused therapeutic strategies (Provan and Newland 2015; Sandal et al., 2021; Provan and Semple 2022; Liu et al., 2023).

Recent investigations have identified Beclin-1 as a crucial regulator of autophagy in the context of ITP pathogenesis (Liu and Mei 2018; Xu and Qin 2019; El-Sayed et al., 2025). Autophagy, the process of cellular disintegration and recycling, is essential for influencing immunological responses, maintaining platelet homeostasis, and regulating megakaryocyte development (Yu et al., 2018). In individuals with ITP, autophagy dysregulation has been linked to heightened platelet apoptosis and alterations in immunological tolerance (Wang et al., 2019). Beclin-1, an essential element of the autophagy-associated phosphatidylinositol 3-kinase (PI3K) complex. It initiates the creation of autophagic vesicles and participates in the functionality of immune cells. Targeting Beclin-1 may facilitate the restoration of autophagy equilibrium, diminish platelet degradation, and unveil novel therapeutic avenues for the management of ITP (Liu et al., 2021; Kamal et al., 2023; Cao et al., 2024).

Computational structure-based virtual screening (SBVS) has emerged as an essential technique for identifying potential drug candidates through molecular docking, molecular dynamics simulations, and in silico pharmacokinetic profiling. This approach facilitates the rapid evaluation of large chemical libraries to identify compounds with high affinity for Beclin-1, thus contributing to the development of novel therapeutics. Integrating computational drug discovery with experimental validation enables researchers to expedite the identification of lead compounds that modulate Beclin-1 activity, thereby offering a more precise treatment strategy for ITP with improved efficacy and safety (Li and Shah 2017; Wang et al., 2020; Bhunia et al., 2021).

This study aims to employ a computational SBVS approach to identify novel small-molecule inhibitors or modulators of Beclin-1. Potential drug candidates were discovered as effective therapy agents for ITP by molecular docking, binding affinity predictions, and pharmacokinetic studies.

2. Materials and Methods

2.1 Generation of pharmacophore model

The crystal structure of the Beclin-1 protein was obtained from the PDB database (PDB ID: 4DDP) and subjected to the Maestro workspace for preparation (Maestro 2020). The structure was optimized and minimized to predict the binding sites of the protein by utilizing the sitemap tool (Ge et al., 2024). The predicted binding site residues were then used to develop pharmacophore models by using Phase (Moumbock et al., 2021). During generation of pharmacophore model, the receptor cavity residues were also selected.

2.2 Virtual screening

The virtual screening was conducted by using the Enamine-AI-Enabled-TF library containing 1520 compounds. Before proceeding towards the screening, the library was processed by Phase to prepare compounds by increasing the chemical space search during which 50 conformers for each compound were generated by Epik (Johnston et al., 2023) at pH 7.0. Additionally, the tautomeric states were removed, and then virtual screening was conducted by utilizing the developed pharmacophore model. The virtual screening results were analysed based on phase screen scores which is combination of root mean square deviation (RMSD) site matching, vector alignment, and volume scores. To select the hit compounds, a threshold of 1.0 phase screen score was used.

2.3 Molecular docking

The hit compounds retrieved from virtual screening were docked against the Beclin-1 receptor. The crystal structure of Beclin-1 was imported and processed by Protein Preparation Wizard in the Maestro tool (Schrödinger 2017). In the pre-process step, bond orders were assigned, polar hydrogens were added, and zero bond orders were created for metals. Water molecules were removed beyond 5 Å, and het atom states were generated using Epik at pH 7.0. In the next step, the hydrogen bonds were optimized using PROPKA at pH 7.0 (Kim et al., 2013). and the energy of the structure was minimized using the OPLS forcefield (Jeong et al., 2025). To perform the site-specific docking, a 3D grid was generated at the predicted binding site residues. The X, Y, Z coordinates were 44.23, 5.24, and 14.85, respectively. After protein preparation, the hits were prepared for docking using LigPrep and docked against the prepared receptor using the SP mode of the glide tool (Friesner et al., 2004). The docking results were analyzed, and drugs were selected based on glide scores.

2.4 Molecular interaction analysis

The screened hits were docked with the Beclin-1 receptor to obtain the binding affinities. The top eight hits were selected based on glide scores, and their molecular interactions were analyzed using the ligand interaction diagram module of Maestro.

2.5 Drug likeness and toxicity analysis

Drug erosion is attributed to toxicity and poor pharmacokinetics issues (Agamah et al., 2020). To overcome these issues, pharmacokinetic profiles of the selected compounds were predicted using the OSIRIS Property Explorer tool (Ejeh et al., 2022). Various properties, like molecular weight, total polar surface area (TPSA), solubility, and LogP values, were predicted. Moreover, the potential toxicity risks of hits were measured. Furthermore, ADMET (absorption, distribution, metabolism, excretion, toxicity) properties were predicted using the QikProp tool (Mali and Chaudhari 2018).

2.6 MD simulation

The stability of protein-ligand complexes was analyzed under the 100 ns MD simulation using GROMACS (Van Der Spoel et al., 2005). The input systems were prepared using the CHARMM-GUI webserver (Jo et al., 2008). During input system preparation, the topology files of the ligands were generated by using CHARMM General Force Field (CGenFF) (Vanommeslaeghe et al., 2012). Similarly, the protein complexes were solvated in a periodic box containing TIP3P water molecules. The systems were neutralized using K⁺ and Cl^- counter ions. After preparing the initial systems, the steric clashes were removed by conducting minimization using the steepest descent method of 5000 steps. Following minimization, the system equilibration was done at 50000 and 100000 steps using NVT and NPT ensembles, respectively. After equilibration, the production run started with Parrinello-Rehman algorithm (Martoňák et al., 2003) and Berendson thermostat (Lemak and Balabaev 1994) to maintain constant temperature at 310 K and pressure at 1 atm. The bond lengths of hydrogen bonds were kept at ideal lengths while the non-bonded interactions were computed using the Verlet algorithm (Grubmüller et al., 1991). Further, the electrostatic interactions were computed by using the particle mesh Ewald approach (Darden et al., 1993). The systems were relaxed by applying the LINCS algorithm (Hess et al., 1997). The MD trajectory was stored at each 50-ps time interval, and simulation analysis was conducted using the GROMACS commands.

3. Results

3.1 Pharmacophore modeling

The crystal structure of Beclin-1 was imported into the Maestro workspace and optimized for the generation of pharmacophore models. A total of seven feature-based pharmacophore models were generated, including R1978, D1013, D1037, A600, A513, N1601, and N1591, as shown in Fig. 1(a-b).

Pharmacophore Model. (a) The predicted pocket of Beclin-1, along with the developed model. (b) The representation of the receptor cavity and the seven features model.

3.2 Virtual screening

The generated pharmacophore hypothesis was used to conduct the virtual screening of the Enamine-AI-Enabled-TF Library. The compounds that matched at least four pharmacophoric features were selected as hits during screening. The phase screen score was used to rank the hits, which is a combination of RMSD matching, vector alignments, and volume scores. The vector score in the range of -1.0 to 1.0 shows better alignment, while the volume reference range is 0.0-1.0, with high scores showing greater overlaps. A phase screen score of 1.0 was used to select the potential hits (Table 1).

Table 1. The screened hits, along with alignment, vector, volume, and phase screen scores.

Catalog ID	Align score	Vector score	Volume score	Phase screen score
Z9255350351	0.675	0.742	0.303	1.264
Z9255349763	0.534	0.744	0.226	1.233
Z9255357155	0.929	0.814	0.269	1.205
Z9255348263	0.672	0.7	0.281	1.202
Z9255354939	0.667	0.677	0.291	1.19
Z9255351313	0.713	0.696	0.23	1.133
Z4129214463	0.834	0.71	0.261	1.132
Z9255355752	0.76	0.643	0.297	1.129
Z9255355511	0.783	0.694	0.245	1.119
Z9255349629	0.625	0.643	0.239	1.118
Z9255354367	0.866	0.676	0.293	1.117
Z9255351313	0.642	0.734	0.151	1.115
Z9255354564	0.998	0.707	0.29	1.114
Z9255348263	0.664	0.649	0.238	1.11
Z4129214463	0.544	0.654	0.194	1.11
Z9255354311	0.651	0.691	0.188	1.106
Z9255356311	0.876	0.7	0.258	1.102
Z1552029202	0.567	0.669	0.179	1.102
Z9255353029	0.958	0.841	0.149	1.099
Z9255353624	0.947	0.678	0.307	1.099
Z9255353624	0.947	0.678	0.307	1.099
Z9255351338	0.876	0.68	0.27	1.095
Z9255351380	0.776	0.739	0.17	1.092
Z9255350335	0.813	0.661	0.261	1.092
Z9255356039	0.624	0.643	0.211	1.09
Z9255355335	0.903	0.639	0.317	1.088
Z9255354895	0.839	0.655	0.272	1.086
Z9255348238	0.749	0.674	0.218	1.085
Z9255354367	0.919	0.682	0.276	1.084
Z9255355511	0.685	0.659	0.209	1.084
Z9255354924	0.469	0.642	0.16	1.084
Z9255356311	0.783	0.672	0.231	1.083
Z9255354971	0.809	0.698	0.213	1.082
Z9255356893	0.621	0.601	0.243	1.081
Z9255353624	1	0.698	0.291	1.081
Z9255351694	0.799	0.621	0.285	1.08
Z9255353203	0.661	0.701	0.153	1.078
Z9255350375	0.702	0.694	0.174	1.078
Z9255350335	0.955	0.686	0.281	1.078
Z9255350970	0.879	0.693	0.242	1.077
Z9255353098	0.86	0.691	0.233	1.075
Z9255349540	0.874	0.664	0.264	1.073
Z9255349008	0.854	0.66	0.26	1.073
Z1672440949	0.752	0.66	0.219	1.072
Z7698810754	0.758	0.701	0.181	1.072
Z1447350407	0.769	0.725	0.16	1.071
Z3984516772	0.704	0.701	0.157	1.066
Z4145361099	0.565	0.62	0.191	1.065
Z9255354723	0.666	0.667	0.176	1.065
Z9255349540	0.837	0.7	0.206	1.065
Z9255352080	0.551	0.621	0.185	1.065
Z9255349532	0.869	0.643	0.275	1.064
Z9255354969	0.632	0.693	0.137	1.064
Z9255354367	0.91	0.654	0.28	1.064
Z2953653517	0.721	0.674	0.186	1.063
Z9255353469	0.749	0.656	0.214	1.063
Z9255349517	0.902	0.667	0.262	1.062
Z9255349532	0.869	0.643	0.272	1.062
Z1736569382	0.652	0.665	0.168	1.06
Z9255354668	0.878	0.681	0.236	1.06
Z9255354723	0.666	0.667	0.168	1.058
Z9255352444	0.803	0.723	0.161	1.056
Z9255354311	0.692	0.655	0.187	1.056
Z9255352059	0.714	0.579	0.27	1.055
Z9255355219	0.91	0.653	0.272	1.054
Z9255352526	0.976	0.701	0.25	1.053
Z737412348	0.874	0.736	0.172	1.052
Z9255355015	0.808	0.631	0.251	1.052
Z9255355481	0.48	0.591	0.182	1.052
Z9255349897	0.928	0.717	0.213	1.052
Z9255348323	0.608	0.603	0.207	1.052
Z240714930	0.875	0.736	0.171	1.051
Z9255351318	1.131	0.709	0.308	1.049
Z9255350910	0.501	0.556	0.218	1.048
Z9255355469	0.762	0.664	0.195	1.047
Z9255349269	0.934	0.736	0.19	1.046
Z9255354786	0.822	0.648	0.232	1.045
Z737487624	0.871	0.734	0.166	1.045
Z9255350375	0.712	0.681	0.158	1.045
Z9255354085	0.898	0.64	0.27	1.044
Z2221991731	0.697	0.707	0.125	1.043
Z9255350375	0.673	0.656	0.165	1.041
Z9255355092	1.008	0.67	0.26	1.041
Z9255354786	0.822	0.648	0.227	1.04
Z9255351642	0.863	0.579	0.312	1.04
Z9255350570	1.052	0.706	0.265	1.039
Z9255349653	0.691	0.704	0.121	1.039
Z9255349405	0.876	0.685	0.209	1.038
Z9255353550	0.834	0.641	0.236	1.038
Z9255349318	0.753	0.568	0.279	1.038
Z1256755459	0.786	0.66	0.199	1.038
Z9255349653	0.691	0.704	0.12	1.038
Z9255354290	0.777	0.721	0.133	1.036
Z9255350375	0.703	0.679	0.147	1.036
Z9255353203	0.662	0.677	0.134	1.035
Z9255350839	0.778	0.679	0.173	1.034
Z9255349100	0.833	0.586	0.287	1.034
Z9255351847	0.934	0.666	0.247	1.034
Z2112679674	0.925	0.593	0.282	1.032
Z9255354939	0.867	0.633	0.25	1.03
Z9255353075	0.959	0.684	0.237	1.03
Z4351152748	0.79	0.663	0.188	1.029
Z9255356840	0.667	0.662	0.144	1.028
Z9255352766	1.006	0.695	0.244	1.028
Z9255350881	0.601	0.553	0.231	1.028
Z8083512386	0.777	0.652	0.193	1.028
Z4476009392	0.747	0.662	0.172	1.028
Z9255353186	0.752	0.638	0.195	1.025
Z9255352819	0.964	0.642	0.275	1.025
Z5017162459	0.834	0.641	0.222	1.024
Z9255356311	0.966	0.668	0.249	1.023
Z9255350898	0.831	0.641	0.22	1.023
Z1410138823	0.834	0.641	0.22	1.022
Z9255355596	0.834	0.641	0.22	1.022
Z9255350977	0.408	0.515	0.209	1.021
Z7837265532	0.786	0.697	0.144	1.021
Z9255355188	0.692	0.606	0.201	1.021
Z1499834521	0.79	0.651	0.192	1.021
Z3561278990	0.435	0.544	0.186	1.02
Z9255352927	0.766	0.639	0.194	1.02
Z9255351165	0.708	0.7	0.112	1.02
Z9255351165	0.708	0.7	0.112	1.02
Z9255351165	0.708	0.7	0.112	1.02
Z9255351165	0.708	0.7	0.112	1.02
Z9255356840	0.849	0.733	0.131	1.019
Z9255355335	0.992	0.633	0.29	1.018
Z9255351681	0.801	0.559	0.285	1.018
Z9255355596	0.834	0.641	0.214	1.016
Z1594062072	0.79	0.672	0.167	1.016
Z9255353962	0.89	0.644	0.234	1.016
Z9255349269	0.956	0.713	0.192	1.015
Z9255353091	0.908	0.652	0.232	1.014
Z9255353509	0.533	0.557	0.193	1.014
Z9255357077	0.715	0.515	0.293	1.014
Z9255353509	0.533	0.557	0.192	1.014
Z9255353186	0.528	0.468	0.279	1.013
Z9255351375	1.058	0.693	0.255	1.013
Z4042414842	0.856	0.707	0.153	1.012
Z8317858793	0.993	0.673	0.244	1.012
Z9255350831	0.701	0.631	0.17	1.012
Z4042420282	0.837	0.7	0.153	1.012
Z9255355525	0.847	0.664	0.192	1.011
Z9255355408	0.917	0.713	0.172	1.011
Z9255352421	0.837	0.691	0.161	1.011
Z1128926614	0.81	0.71	0.132	1.011
Z9255354312	0.949	0.655	0.242	1.011
Z9255354924	0.719	0.616	0.19	1.011
Z9255355612	0.967	0.715	0.19	1.011
Z9255354009	0.714	0.605	0.2	1.01
Z9255351589	0.837	0.691	0.16	1.01
Z9255351810	0.842	0.628	0.224	1.01
Z9255356840	0.825	0.663	0.184	1.01
Z9255352421	0.837	0.691	0.159	1.01
Z9255351589	0.837	0.691	0.159	1.01
Z2613685214	0.822	0.672	0.172	1.009
Z9255355469	0.863	0.68	0.179	1.008
Z9255353522	0.937	0.63	0.26	1.008
Z2613685214	0.825	0.672	0.173	1.008
Z9255352697	0.641	0.61	0.167	1.008
Z9255348343	0.933	0.617	0.27	1.007
Z9255354095	0.847	0.657	0.193	1.005
Z9255350112	0.849	0.678	0.172	1.005
Z9255350722	0.969	0.631	0.269	1.005
Z9255355015	0.837	0.691	0.154	1.005
Z4167476424	0.857	0.652	0.201	1.004
Z8083512386	0.76	0.602	0.212	1.003
Z9255353025	0.8	0.544	0.286	1.003
Z9255354213	0.813	0.607	0.228	1.003
Z9255354213	0.813	0.607	0.228	1.003
Z7872360283	0.752	0.629	0.18	1.001

3.3 Molecular docking

The compounds screened by virtual screening were prepared and there docking study was conducted against the Beclin-1 protein (Kumari et al., 2023). The binding affinities of all docked compounds were analysed and then the top ten compounds were selected for further analysis (Table 2). The binding affinities of the selected hits were in the range of -5.766 to -5.103 kcal/mol. The docking scores of the selected compounds suggested that these have potential for inhibiting the function of the Beclin-1 protein.

Table 2. The glide scores of the docked compounds against Beclin-1.

Sr.	Catalog ID	Glide score (kcal/mol)
1	Z1672440949	-5.766
2	Z9255349517	-5.501
3	Z9255355511	-5.382
4	Z7837265532	-5.337
5	Z9255356311	-5.407
6	Z9255355469	-5.191
7	Z9255355525	-5.127
8	Z9255355511	-5.103

3.4 Post docking analysis

The molecular interactions of the selected hits with Beclin-1 binding sites were analyzed using the Ligand interaction diagram module of Maestro. The interactions mainly involve hydrogen bonds, pi-cation interactions, pi-pi stacking, salt bridges, and hydrophobic interactions. These interactions play a key role in determining the binding affinities of candidate compounds. The overall stability of the protein-ligand complex is dependent on these interactions (Thillainayagam et al., 2018). The binding affinities and molecular interactions of selected hits have been shown in Table 3. The summary of residues forming hydrogen bonds and hydrophobic interactions has also been shown in Table 3. In the interaction diagrams, hydrogen bonds are shown with purple arrows, pi-cation with red arrows, while hydrophobic interactions are shown in pale yellow spheres.

3.5 ADMET analysis

The toxicity analysis was performed using the OSIRIS Property Explorer, which predicted mutagenic, carcinogenic, and reproductive toxicity risks. Compounds were considered suitable for further development if they passed the standard toxicity thresholds used in preclinical screening, ensuring a lower risk of adverse effects in vivo. In pharmacokinetic properties, molecular weight, LogP, TPSA, and solubility (LogS) of the hits were estimated. Molecular weight helps in determining the easy distribution of the drug within the cells, so a hit compound with a lower molecular weight can easily dissolve in the body. Similarly, hydrophilicity indicates the absorption of a compound, which is determined by calculating LogP values. LogP > 5 indicates poor absorption of compounds. Another parameter is TPSA, which is related to the hydrogen bonding ability of a compound and is a good predictor of bioavailability (Husain et al., 2016). Compounds having a TPSA value<160 Å² have good oral bioavailability (Gogoi et al., 2021). Lastly, solubility helps in measuring the ability of a compound to dissolve in the solvent. The pharmacokinetic profiles of the hits showed that the hits met the threshold values for M.W < 500, LogP < 5, TPSA < 160 Å², and LogS < -5 (Table 4). In addition to these parameters, the drug-likeness and drug scores of the hits were also predicted. Positive value for drug-likeness shows that the compound has some common structural features with known drugs, while drug scores are the combination of drug-likeness, solubility, molecular weight, hydrophilicity, and toxicity risks. The higher the drug score for a compound, the higher the potential of that compound to be developed into a medication (Behrouz et al., 2019). Moreover, the drug toxicity analysis was performed to evaluate the potential risks for hits to be mutagenic, tumorigenic, irritant, with reproductive effects. The analysis revealed that all hits passed the toxicity risk test (Table 4). Furthermore, ADMET properties were also predicted by the QikProp tool, as shown in Table 5. The values for all the parameters were in the cutoff range except for QPPCaco values. Only two compounds met this criterion (“QPlogHERG” (<-5), “QPlogPo/w” (-2.0 to 6.5), “QPlogBB” (-3.0 to 1.2), “QPPCaco” (<25 poor, >500 great), and “QPlogKhsa” (-1.5 to 1.5)), so these were selected for further analysis.

Table 4. The physicochemical and toxicity profiles of the selected hits against Beclin-1.

	Pharmacokinetic Properties						Toxicity Profiles
Catalog ID	MW	cLogP	TPSA	LogS	Druglikeness	Drug score	Mutagenic	Tumorigenic	Irritant	Reproductive effect
Z1672440949	408	3.68	102.5	-1.27	-2.21	0.43	Passed	Passed	Passed	Passed
Z9255349517	383	-0.83	97.94	-0.71	7.57	0.89	Passed	Passed	Passed	Passed
Z9255355511	469	3.01	105.5	-5.23	2.38	0.52	Passed	Passed	Passed	Passed
Z7837265532	305	0.73	85.32	-3.17	4.86	0.88	Passed	Passed	Passed	Passed
Z9255356311	406	1.37	96.69	-2.09	4.31	0.84	Passed	Passed	Passed	Passed
Z9255355469	418	2.3	86	-4.21	2.52	0.68	Passed	Passed	Passed	Passed
Z9255355525	322	0.58	107.9	-3.51	-1.16	0.53	Passed	Passed	Passed	Passed
Z9255355511	469	3.01	105.5	-5.23	2.38	0.52	Passed	Passed	Passed	Passed

Table 5. The ADMET profiles of the selected hits.

Catalog ID	QPlogPo/w	QPlogHERG	QPPCaco	QPlogBB	QPlogKhsa
Z1672440949	3.893	-4.832	495.192	-0.999	0.561
Z9255349517	1.684	-3.278	371.457	-1.259	-0.694
Z9255355511	2.779	-3.553	68.083	-1.497	-0.104
Z7837265532	2.273	-5.482	326.518	-1.083	-0.126
Z9255356311	2.165	-4.247	593.303	-0.731	-0.152
Z9255355469	4.266	-5.175	619.147	-0.696	0.448
Z9255355525	1.289	-4.991	200.95	-1.293	-0.23
Z9255355511	2.046	-1.879	90.269	-1.144	-0.302

3.6 Binding pose alignment

After the selection of two compounds, their binding modes in the receptor binding pocket cavity were observed. Both ligands were superimposed on each other, which revealed that the hit compound occupied the same space in the Beclin-1, indicating the accuracy of the docking protocol (Fig. 2).

The alignment of binding modes of selected hits in the Beclin-1 binding pocket. (a) Z9255356311, (b) Z9255355469.

3.7 MD simulation

The stability of selected hit compounds with the Beclin-1 receptor was estimated by conducting 100 simulations, and the MD trajectory was analyzed to calculate RMSD, root mean square fluctuation (RMSF), radius of gyration, and hydrogen bonding analysis.

3.7.1 RMSD

The RMSD of carbon alpha atoms of protein and ligand atoms was calculated to estimate the protein-ligand complex stability (Sargsyan et al., 2017). Protein RMSD shows the deviation of protein structure from its initial confirmation, while the ligand RMSD values show how strongly or loosely a ligand is bound to the protein. In Z9255356311-complex, the protein RMSD fluctuates around 0.3-0.5 nm, indicating moderate structural stability with some flexibility. The ligand RMSD starts low (∼0.1 nm) and gradually increases to around 0.25 nm, suggesting that the ligand remains relatively stable (Fig. 3(a)). Similarly, in the Z9255355469 complex, protein RMSD fluctuates between approximately 0.2-0.21 nm, showing more structural stability. While the ligand RMSD varies between 0.1 and 0.25 nm, indicating moderate ligand mobility but still generally stable binding (Fig. 3(b)).

RMSD analysis of protein and ligand atoms during 100 ns molecular dynamics simulations for selected protein-ligand complexes. (a) Z9255356311, (b) Z9255355469.

3.7.2 RMSF

RMSF measures the average deviation of each residue from its mean position during the MD simulation, providing insight into the flexibility and mobility of different regions of the protein (Martínez 2015). The RMSF analysis of four complexes was conducted as displayed in Fig. 4. The figure shows that all complexes display several peaks, indicating regions of high and low flexibility, respectively. The highest RMSF values (peaks) are typically observed at the N- and C-termini, as well as at specific internal segments, likely corresponding to loop or unstructured regions. Most of the protein residues exhibit RMSF values between 0.1 and 0.2 nm, suggesting these regions are relatively stable and less flexible. The RMSF profiles reveal that certain regions of the protein, particularly around residue 120 and the terminal residues, are consistently more flexible across all four systems.

Root Mean Square Fluctuation (RMSF) profiles of Cα atoms for four different protein systems. (a) Z9255356311, (b) Z9255355469.

3.7.3 Radius of gyration

The radius of gyration was calculated for both complexes. It shows the compactness of the system during simulation; the lower the Rg values, the more compact the system will be. The Rg values for complex-Z9255356311 were between 1.63 and 1.75 nm, while the average value was 1.68 Fig. 5. The Rg values for Z9255355469-complex were slightly lower than those of the first complex, as the values were between 1.62 and 1.67 nm, with an average value of 1.65 nm. The Rg values did not show too many deviations, indicating that the complexes remained compact during simulation.

The evolution of Rg values during a 100 ns simulation for both complexes. (a) Z9255356311, (b) Z9255355469.

3.7.4 Hydrogen bonding

The hydrogen bonding between protein and ligand shows the stability of the complex, so it was calculated during simulation. Hydrogen bonding analysis revealed that in complex-Z9255356311, the hydrogen bonds rarely exceed 2, and at some frames, the bonding was not observed Fig. 6(a). While in the second complex, the hydrogen bonds were between 1 and 4 at most of the frames, the number of hydrogen bonds increased after 80 ns, indicating more stability in the last part of the simulation Fig. 6(b).

The estimation of hydrogen bonds in selected complexes during 100 ns simulation. (a) Z9255356311, (b) Z9255355469.

4. Discussion

This research employed a computational SBVS approach to identify novel inhibitors that specifically target Beclin-1, a critical regulator of autophagy associated with ITP (Zufferey et al., 2017; Cooper and Ghanima 2019; Cao et al., 2024). Through the implementation of systematic screening, molecular docking, post-docking analysis, ADMET profiling, and molecular dynamics (MD) simulations, many promising small molecules have been identified as potential therapeutic agents. The findings provide critical information regarding the structural interactions of Beclin-1 inhibitors, including their stability, binding efficiency, and drug-like properties, which contribute to the advancement of targeted therapies for ITP(Katsila et al., 2016; Pathak et al., 2020; Sadybekov and Katritch 2023).

The Beclin-1 structure was obtained from the Protein Data Bank (PDB) and subsequently prepared for virtual screening. The virtual screening process commenced with the creation of an accurate structure-based pharmacophore model, which was subjected to perform virtual screening of Enamine-AI-Enabled-TF library. Initial screening resulted in 170 compounds, followed by refinement through molecular docking studies. The docking results demonstrated that eight compounds exhibited strong binding to Beclin-1, with binding affinities spanning from –5.766 to -5.103 kcal/mol.

An in-depth examination of post-docking interactions was performed to elucidate the binding mechanisms of the top-ranked compounds to Beclin-1. The primary stabilizing forces in protein-ligand complexes consist of hydrogen bonding, van der Waals interactions, pi-pi stacking, and hydrophobic interactions. ADMET analysis was conducted to evaluate the pharmacokinetic and drug-like properties of the selected compounds(Kar and Leszczynski 2020; Daoud et al., 2021). The compounds underwent evaluation focusing on their absorption, distribution, metabolism, excretion, and toxicity characteristics, employing defined cut-off values to ascertain their appropriateness. Two compounds satisfied the selection criteria, demonstrating favorable pharmacokinetic profiles and remaining within acceptable toxicity limits.

MD simulations were conducted to evaluate the conformational stability of selected protein-ligand complexes over a duration of 100 ns (Shukla and Tripathi 2021; Adelusi et al., 2022; Filipe and Loura 2022). The RMSD values of the complexes demonstrated that both compounds exhibited stable interactions during the entire simulation period. Further insights into protein-ligand dynamics were obtained through root mean square fluctuation (RMSF) analysis. This analysis indicated that most protein residues exhibited minimal fluctuations, remaining below 0.2 nm, while loop regions demonstrated increased flexibility. This observation suggests that the binding of these ligands did not result in notable structural alterations, thereby supporting their viability as stable inhibitors. The persistence of hydrogen bonds observed in molecular dynamics simulations highlights their critical role in stabilizing protein-ligand complexes, indicating that these compounds may serve as effective Beclin-1 inhibitors.

The identification of potential Beclin-1 inhibitors presents a novel therapeutic strategy for the treatment of ITP. Modulating Beclin-1 activity may enable the restoration of autophagy balance, reduction of platelet destruction, and alleviation of immune dysfunction associated with the disease. Among the three compounds evaluated, Carbamazepine exhibited the most favorable combination of binding affinity, ADMET properties, and dynamic stability, establishing it as a leading candidate for subsequent experimental testing. Future research should focus on conducting in vitro and in vivo studies to validate its efficacy and safety within biological systems.

5. Conclusions

This study utilized a computational, SBVS approach to identify potential Beclin-1 inhibitors for the treatment of ITP. Through a combination of pharmacophore modeling, molecular docking, ADMET profiling, and molecular dynamics simulations, Z9255356311 and Z9255355469 were identified as promising candidates. These compounds exhibited stable interactions with Beclin-1, favorable pharmacokinetic properties, and good drug-likeness, making them strong candidates for further development. The findings suggest that, based on the in silico analysis, Z9255356311 and Z9255355469 can act as potential leads for Beclin-1 inhibition.

CRediT authorship contribution statement

RuiBo Zhang: Conceptualization, methodology, data analysis, writing—original draft, writing—review and editing. Juan Peng: Data curation, molecular docking, validation, writing—review and editing. Lan Ren: Molecular dynamics simulations, result interpretation, writing—review and editing. Jingjing Cheng: Structural analysis, figure preparation, writing—review and editing. YiGang Guo: Data analysis, literature review, writing—review and editing. Kun Zheng: Supervision, project administration, funding acquisition, writing—review and editing.

Declaration of competing interest

The authors declare that they have no competing financial interests or personal relationships that could have influenced the work presented in this paper.

Declaration of Generative AI and AI-assisted technologies in the writing process

Use of Artificial Intelligence (AI)-Assisted Technology for manuscript preparation: The authors confirm that they have used Artificial Intelligence (AI)-Assisted Technology for assisting in the writing or editing of the manuscript or image creations.

References

Adelusi T.I., Oyedele A.-Q.K., Boyenle I.D., Ogunlana A.T., Adeyemi R.O., Ukachi C.D., Idris M.O., Olaoba O.T., Adedotun I.O., Kolawole O.E., Xiaoxing Y., Abdul-Hammed M.. Molecular modeling in drug discovery. Informatics in Medicine Unlocked. 2022;29:100880. https://doi.org/10.1016/j.imu.2022.100880
[Google Scholar]
Agamah F.E., Mazandu G.K., Hassan R., Bope C.D., Thomford N.E., Ghansah A., Chimusa E.R.. Computational/in silico methods in drug target and lead prediction. Brief Bioinform. 2020;21:1663-1675. https://doi.org/10.1093/bib/bbz103
[Google Scholar]
Behrouz S., Soltani Rad M.N., Taghavi Shahraki B., Fathalipour M., Behrouz M., Mirkhani H.. Design, synthesis, and in silico studies of novel eugenyloxy propanol azole derivatives having potent antinociceptive activity and evaluation of their β-adrenoceptor blocking property. Mol Divers. 2019;23:147-164. https://doi.org/10.1007/s11030-018-9867-7
[Google Scholar]
Bhunia S.S., Saxena M., Saxena A.K.. Ligand- and structure-based virtual screening in drug discovery. In: Biophysical and Computational Tools in Drug Discovery. Cham: Springer International Publishing; 2021. p. :281-339. https://doi.org/10.1007/7355_2021_130
[Google Scholar]
Cao Z., Tian K., Ran Y., Zhou H., Zhou L., Ding Y., Tang X.. Beclin-1: A therapeutic target at the intersection of autophagy, immunotherapy, and cancer treatment. Front Immunol. 2024;15:1506426. https://doi.org/10.3389/fimmu.2024.1506426
[Google Scholar]
Cooper N., Ghanima W.. Immune Thrombocytopenia. N Engl J Med. 2019;381:945-955. https://doi.org/10.1056/NEJMcp1810479
[Google Scholar]
Daoud N.E., Borah P., Deb P.K., Venugopala K.N., Hourani W., Alzweiri M., Bardaweel S.K., Tiwari V.. ADMET profiling in drug discovery and development: perspectives of in silico, in vitro and integrated approaches. Curr Drug Metab. 2021;22:503-522. https://doi.org/10.2174/1389200222666210705122913
[Google Scholar]
Darden T., York D., Pedersen L.J.T.J.O.C.P.. Particle mesh Ewald: An N⋅ log (N) method for Ewald sums in large systems. 98 1993:10089-10092.
[Google Scholar]
Ejeh S., Uzairu A., Shallangwa G.A., Abechi S.E., Ibrahim M.T.. In silico design and pharmacokinetics investigation of some novel hepatitis C virus NS5B inhibitors: Pharmacoinformatics approach. Bull Natl Res Cent. 2022;46 https://doi.org/10.1186/s42269-022-00796-y
[Google Scholar]
El-Sayed N.E.T., Badr Eldin N.M., Elrahman Sweilam M.A., Ahmed A.Y., Elhawary E.E.. The role of autophagy-related markers Beclin-1, LC3, and P62 in the pathogenesis of immune thrombocytopenia in children. J Thromb Haemost. 2025;23:3362-3369. https://doi.org/10.1016/j.jtha.2025.06.034
[Google Scholar]
Filipe H.A.L., Loura L.M.S.. Molecular dynamics simulations: Advances and applications. Molecules. 2022;27:2105. https://doi.org/10.3390/molecules27072105
[Google Scholar]
Friesner R.A., Banks J.L., Murphy R.B., Halgren T.A., Klicic J.J., Mainz D.T., Repasky M.P., Knoll E.H., Shelley M., Perry J.K., Shaw D.E., Francis P., Shenkin P.S.. Glide: A new approach for rapid, accurate docking and scoring 1 Method and assessment of docking accuracy. J Med Chem. 2004;47:1739-1749. https://doi.org/10.1021/jm0306430
[Google Scholar]
Ge Y., Pande V., Seierstad M.J., Damm-Ganamet K.L.. Exploring the application of SiteMap and site finder for focused cryptic pocket identification. J Phys Chem B. 2024;128:6233-6245. https://doi.org/10.1021/acs.jpcb.4c00664
[Google Scholar]
Gogoi P., Shakya A., Ghosh S.K., Gogoi N., Gahtori P., Singh N., Bhattacharyya D.R., Singh U.P., Bhat H.R.. In silico study, synthesis, and evaluation of the antimalarial activity of hybrid dimethoxy pyrazole 1,3,5‐triazine derivatives. J Biochem & Molecular Tox. 2021;35 https://doi.org/10.1002/jbt.22682
[Google Scholar]
Grubmüller H., Heller H., Windemuth A., Schulten K.. Generalized Verlet algorithm for efficient molecular dynamics simulations with long-range interactions. Molecular Simulation. 1991;6:121-142. https://doi.org/10.1080/08927029108022142
[Google Scholar]
Hess B., Bekker H., Berendsen H.J.C., Fraaije J.G.E.M.. LINCS: A linear constraint solver for molecular simulations. J Comput Chem. 1997;18:1463-1472. https://doi.org/10.1002/(SICI)1096-987X(199709)18:12<1463::AID-JCC4>3.0.CO;2-H
[Google Scholar]
Husain A., Ahmad A., Khan S.A., Asif M., Bhutani R., Al-Abbasi F.A.. Synthesis, molecular properties, toxicity and biological evaluation of some new substituted imidazolidine derivatives in search of potent anti-inflammatory agents. Saudi Pharm J. 2016;24:104-114. https://doi.org/10.1016/j.jsps.2015.02.008
[Google Scholar]
Jeong S.W., Kim Y., Lim H.-K., Chan Kim M., Song C.H., Lee S.. Prediction of enthalpy of vaporization for particulate matter through molecular dynamics using OPLS force field. J Aerosol Sci. 2025;188:106595. https://doi.org/10.1016/j.jaerosci.2025.106595
[Google Scholar]
Jo S., Kim T., Iyer V.G., Im W.. CHARMM-GUI: A web-based graphical user interface for CHARMM. J Comput Chem. 2008;29:1859-1865. https://doi.org/10.1002/jcc.20945
[Google Scholar]
Johnston R.C., Yao K., Kaplan Z., Chelliah M., Leswing K., Seekins S., Watts S., Calkins D., Chief Elk J., Jerome S.V., Repasky M.P., Shelley J.C.. Epik: P Ka and protonation state prediction through machine learning. J Chem Theory Comput. 2023;19:2380-2388. https://doi.org/10.1021/acs.jctc.3c00044
[Google Scholar]
Kamal A.M., Nabih N.A., Rakha N.M., Sanad E.F.. Upregulation of necroptosis markers RIPK3/MLKL and their crosstalk with autophagy-related protein Beclin-1 in primary immune thrombocytopenia. Clin Exp Med. 2023;23:447-456. https://doi.org/10.1007/s10238-022-00839-8
[Google Scholar]
Kar S., Leszczynski J.. Open access in silico tools to predict the ADMET profiling of drug candidates. Expert Opin Drug Discov. 2020;15:1473-1487. https://doi.org/10.1080/17460441.2020.1798926
[Google Scholar]
Kashiwagi H., Tomiyama Y.. Pathophysiology and management of primary immune thrombocytopenia. Int J Hematol. 2013;98:24-33. https://doi.org/10.1007/s12185-013-1370-4
[Google Scholar]
Katsila T., Spyroulias G.A., Patrinos G.P., Matsoukas M.T.. Computational approaches in target identification and drug discovery. Comput Struct Biotechnol J. 2016;14:177-184. https://doi.org/10.1016/j.csbj.2016.04.004
[Google Scholar]
Kim M.O., Nichols S.E., Wang Y., McCammon J.A.. Effects of histidine protonation and rotameric states on virtual screening of m. tuberculosis RmlC. J Comput Aided Mol Des. 2013;27:235-246. https://doi.org/10.1007/s10822-013-9643-9
[Google Scholar]
Kumari R., Rathi R., Pathak S.R., Dalal V.. Computational investigation of potent inhibitors against YsxC: Structure-based pharmacophore modeling, molecular docking, molecular dynamics, and binding free energy. J Biomol Struct Dyn. 2023;41:930-941. https://doi.org/10.1080/07391102.2021.2015446
[Google Scholar]
Lemak A.S., Balabaev N.K.. On the Berendsen thermostat. Mol Simul. 1994;13:177-187. https://doi.org/10.1080/08927029408021981
[Google Scholar]
Li J., van der Wal D.E., Zhu G., Xu M., Yougbare I., Ma L., Vadasz B., Carrim N., Grozovsky R., Ruan M., Zhu L., Zeng Q., Tao L., Zhai Z.M., Peng J., Hou M., Leytin V., Freedman J., Hoffmeister K.M., Ni H.. Desialylation is a mechanism of Fc-independent platelet clearance and a therapeutic target in immune thrombocytopenia. Nat Commun. 2015;6:7737. https://doi.org/10.1038/ncomms8737
[Google Scholar]
Li Q., Shah S.. Structure-based virtual screening. In: Protein Bioinformatics. New York, NY: Springer New York; 2017. p. :111-124. https://doi.org/10.1007/978-1-4939-6783-4_5
[Google Scholar]
Liu S.‐Y, Zhang X.‐M., Sun R.‐J., Zhu J.‐J, Yuan D., Shan N.‐N. Abnormal expression of autophagy‐related proteins in immune thrombocytopenia. Scand J Immunol. 2021;93 https://doi.org/10.1111/sji.12992
[Google Scholar]
Liu X.G., Hou Y., Hou M.. How we treat primary immune thrombocytopenia in adults. J Hematol Oncol. 2023;16:4. https://doi.org/10.1186/s13045-023-01401-z
[Google Scholar]
Liu Z., Mei T.. Immune thrombocytopenia induces autophagy and suppresses apoptosis in megakaryocytes. Mol Med Rep. 2018;18:4016-4022. https://doi.org/10.3892/mmr.2018.9373
[Google Scholar]
Maestro S.J.S.. Llc, New York, Ny. Maestro. 2020;2020:682.
[Google Scholar]
Mali S.N., Chaudhari H.K.. Computational studies on Imidazo[1,2-a] Pyridine-3-carboxamide analogues as antimycobacterial agents: Common pharmacophore generation, atom-based 3D-QSAR, molecular dynamics simulation, QikProp, molecular docking and prime MMGBSA approaches. Pharmsci. 2018;5:12-23. https://doi.org/10.2174/1874844901805010012
[Google Scholar]
Martínez L.. Automatic identification of mobile and rigid substructures in molecular dynamics simulations and fractional structural fluctuation analysis. PLoS One. 2015;10:e0119264. https://doi.org/10.1371/journal.pone.0119264
[Google Scholar]
Martonák R., Laio A., Parrinello M.. Predicting crystal structures: The Parrinello-Rahman method revisited. Phys Rev Lett. 2003;90:075503. https://doi.org/10.1103/PhysRevLett.90.075503
[Google Scholar]
Moumbock A.F.A., Li J., Tran H.T.T., Hinkelmann R., Lamy E., Jessen H.J., Günther S.. ePharmaLib: A versatile library of e-Pharmacophores to address small-molecule (poly-)pharmacology. J Chem Inf Model. 2021;61:3659-3666. https://doi.org/10.1021/acs.jcim.1c00135
[Google Scholar]
Pathak R.K., Singh D.B., Sagar M., Baunthiyal M., Kumar A.. Computational approaches in drug discovery and design. In: Computer-Aided Drug Design. Singapore: Springer; 2020. p. :1-21. https://doi.org/10.1007/978-981-15-6815-2_1
[Google Scholar]
Provan D., Newland A.C.. Current management of primary immune thrombocytopenia. Adv Ther. 2015;32:875-887. https://doi.org/10.1007/s12325-015-0251-z
[Google Scholar]
Provan D., Semple J.W.. Recent advances in the mechanisms and treatment of immune thrombocytopenia. EBioMedicine. 2022;76:103820. https://doi.org/10.1016/j.ebiom.2022.103820
[Google Scholar]
Sadybekov A.V., Katritch V.. Computational approaches streamlining drug discovery. Nature. 2023;616:673-685. https://doi.org/10.1038/s41586-023-05905-z
[Google Scholar]
Sandal R., Mishra K., Jandial A., Sahu K.K., Siddiqui A.D.. Update on diagnosis and treatment of immune thrombocytopenia. Expert Rev Clin Pharmacol. 2021;14:553-568. https://doi.org/10.1080/17512433.2021.1903315
[Google Scholar]
Sargsyan K., Grauffel C., Lim C.. How molecular size impacts RMSD applications in molecular dynamics simulations. J Chem Theory Comput. 2017;13:1518-1524. https://doi.org/10.1021/acs.jctc.7b00028
[Google Scholar]
Schrödinger L.J.S.S.. Version 2, 2017-2011. New York, NY: Schrödinger, LLC; 2017.
Shukla R., Tripathi T.. Molecular dynamics simulation in drug discovery: opportunities and challenges. In: Innovations and Implementations of Computer Aided Drug Discovery Strategies in Rational Drug Design. Singapore: Springer Singapore; 2021. p. :295-316. https://doi.org/10.1007/978-981-15-8936-2_12
[Google Scholar]
Thillainayagam M., Malathi K., Ramaiah S.. In-silico molecular docking and simulation studies on novel chalcone and flavone hybrid derivatives with 1, 2, 3-triazole linkage as vital inhibitors of Plasmodium falciparum dihydroorotate dehydrogenase. J Biomol Struct Dyn. 2018;36:3993-4009. https://doi.org/10.1080/07391102.2017.1404935
[Google Scholar]
Van Der Spoel D., Lindahl E., Hess B., Groenhof G., Mark A.E., Berendsen H.J.. GROMACS: Fast, flexible, and free. J Comput Chem. 2005;26:1701-1718. https://doi.org/10.1002/jcc.20291
[Google Scholar]
Vanommeslaeghe K., MacKerell A.D.. Automation of the CHARMM general force field (CGenFF) i: Bond perception and atom typing. J Chem Inf Model. 2012;52:3144-3154. https://doi.org/10.1021/ci300363c
[Google Scholar]
Wang C.Y., Ma S., Bi S.J., Su L., Huang S.Y., Miao J.Y., Ma C.H., Gao C.J., Hou M., Peng J.. Enhancing autophagy protects platelets in immune thrombocytopenia patients. Ann Transl Med. 2019;7:134. https://doi.org/10.21037/atm.2019.03.04
[Google Scholar]
Wang Z., Sun H., Shen C., Hu X., Gao J., Li D., Cao D., Hou T.. Combined strategies in structure-based virtual screening. Phys Chem Chem Phys. 2020;22:3149-3159. https://doi.org/10.1039/c9cp06303j
[Google Scholar]
Xu H.D., Qin Z.H.. Beclin 1, Bcl-2 and Autophagy. Adv Exp Med Biol. 2019;1206:109-126. https://doi.org/10.1007/978-981-15-0602-4_5
[Google Scholar]
Yu L., Chen Y., Tooze S.A.. Autophagy pathway: Cellular and molecular mechanisms. Autophagy. 2018;14:207-215. https://doi.org/10.1080/15548627.2017.1378838
[Google Scholar]
Zhou H., Hou Y., Liu X., Qiu J., Feng Q., Wang Y., Zhang X., Min Y., Shao L., Liu X., Li G., Li L., Yang L., Xu S., Ni H., Peng J., Hou M.. Low-dose decitabine promotes megakaryocyte maturation and platelet production in healthy controls and immune thrombocytopenia. Thromb Haemost. 2015;113:1021-1034. https://doi.org/10.1160/TH14-04-0342
[Google Scholar]
Zufferey A., Kapur R., Semple J.W.. Pathogenesis and therapeutic mechanisms in immune thrombocytopenia (ITP) J Clin Med. 2017;6:16. https://doi.org/10.3390/jcm6020016
[Google Scholar]

Show Sections

[1] Adelusi T.I., Oyedele A.-Q.K., Boyenle I.D., Ogunlana A.T., Adeyemi R.O., Ukachi C.D., Idris M.O., Olaoba O.T., Adedotun I.O., Kolawole O.E., Xiaoxing Y., Abdul-Hammed M.. Molecular modeling in drug discovery. Informatics in Medicine Unlocked. 2022;29:100880. https://doi.org/10.1016/j.imu.2022.100880
[Google Scholar]

[2] Agamah F.E., Mazandu G.K., Hassan R., Bope C.D., Thomford N.E., Ghansah A., Chimusa E.R.. Computational/in silico methods in drug target and lead prediction. Brief Bioinform. 2020;21:1663-1675. https://doi.org/10.1093/bib/bbz103
[Google Scholar]

[3] Behrouz S., Soltani Rad M.N., Taghavi Shahraki B., Fathalipour M., Behrouz M., Mirkhani H.. Design, synthesis, and in silico studies of novel eugenyloxy propanol azole derivatives having potent antinociceptive activity and evaluation of their β-adrenoceptor blocking property. Mol Divers. 2019;23:147-164. https://doi.org/10.1007/s11030-018-9867-7
[Google Scholar]

[4] Bhunia S.S., Saxena M., Saxena A.K.. Ligand- and structure-based virtual screening in drug discovery. In: Biophysical and Computational Tools in Drug Discovery. Cham: Springer International Publishing; 2021. p. :281-339. https://doi.org/10.1007/7355_2021_130
[Google Scholar]

[5] Cao Z., Tian K., Ran Y., Zhou H., Zhou L., Ding Y., Tang X.. Beclin-1: A therapeutic target at the intersection of autophagy, immunotherapy, and cancer treatment. Front Immunol. 2024;15:1506426. https://doi.org/10.3389/fimmu.2024.1506426
[Google Scholar]

[6] Cooper N., Ghanima W.. Immune Thrombocytopenia. N Engl J Med. 2019;381:945-955. https://doi.org/10.1056/NEJMcp1810479
[Google Scholar]

[7] Daoud N.E., Borah P., Deb P.K., Venugopala K.N., Hourani W., Alzweiri M., Bardaweel S.K., Tiwari V.. ADMET profiling in drug discovery and development: perspectives of in silico, in vitro and integrated approaches. Curr Drug Metab. 2021;22:503-522. https://doi.org/10.2174/1389200222666210705122913
[Google Scholar]

[8] Darden T., York D., Pedersen L.J.T.J.O.C.P.. Particle mesh Ewald: An N⋅ log (N) method for Ewald sums in large systems. 98 1993:10089-10092.
[Google Scholar]

[9] Ejeh S., Uzairu A., Shallangwa G.A., Abechi S.E., Ibrahim M.T.. In silico design and pharmacokinetics investigation of some novel hepatitis C virus NS5B inhibitors: Pharmacoinformatics approach. Bull Natl Res Cent. 2022;46 https://doi.org/10.1186/s42269-022-00796-y
[Google Scholar]

[10] El-Sayed N.E.T., Badr Eldin N.M., Elrahman Sweilam M.A., Ahmed A.Y., Elhawary E.E.. The role of autophagy-related markers Beclin-1, LC3, and P62 in the pathogenesis of immune thrombocytopenia in children. J Thromb Haemost. 2025;23:3362-3369. https://doi.org/10.1016/j.jtha.2025.06.034
[Google Scholar]

[11] Filipe H.A.L., Loura L.M.S.. Molecular dynamics simulations: Advances and applications. Molecules. 2022;27:2105. https://doi.org/10.3390/molecules27072105
[Google Scholar]

[12] Friesner R.A., Banks J.L., Murphy R.B., Halgren T.A., Klicic J.J., Mainz D.T., Repasky M.P., Knoll E.H., Shelley M., Perry J.K., Shaw D.E., Francis P., Shenkin P.S.. Glide: A new approach for rapid, accurate docking and scoring 1 Method and assessment of docking accuracy. J Med Chem. 2004;47:1739-1749. https://doi.org/10.1021/jm0306430
[Google Scholar]

[13] Ge Y., Pande V., Seierstad M.J., Damm-Ganamet K.L.. Exploring the application of SiteMap and site finder for focused cryptic pocket identification. J Phys Chem B. 2024;128:6233-6245. https://doi.org/10.1021/acs.jpcb.4c00664
[Google Scholar]

[14] Gogoi P., Shakya A., Ghosh S.K., Gogoi N., Gahtori P., Singh N., Bhattacharyya D.R., Singh U.P., Bhat H.R.. In silico study, synthesis, and evaluation of the antimalarial activity of hybrid dimethoxy pyrazole 1,3,5‐triazine derivatives. J Biochem & Molecular Tox. 2021;35 https://doi.org/10.1002/jbt.22682
[Google Scholar]

[15] Grubmüller H., Heller H., Windemuth A., Schulten K.. Generalized Verlet algorithm for efficient molecular dynamics simulations with long-range interactions. Molecular Simulation. 1991;6:121-142. https://doi.org/10.1080/08927029108022142
[Google Scholar]

[16] Hess B., Bekker H., Berendsen H.J.C., Fraaije J.G.E.M.. LINCS: A linear constraint solver for molecular simulations. J Comput Chem. 1997;18:1463-1472. https://doi.org/10.1002/(SICI)1096-987X(199709)18:12<1463::AID-JCC4>3.0.CO;2-H
[Google Scholar]

[17] Husain A., Ahmad A., Khan S.A., Asif M., Bhutani R., Al-Abbasi F.A.. Synthesis, molecular properties, toxicity and biological evaluation of some new substituted imidazolidine derivatives in search of potent anti-inflammatory agents. Saudi Pharm J. 2016;24:104-114. https://doi.org/10.1016/j.jsps.2015.02.008
[Google Scholar]

[18] Jeong S.W., Kim Y., Lim H.-K., Chan Kim M., Song C.H., Lee S.. Prediction of enthalpy of vaporization for particulate matter through molecular dynamics using OPLS force field. J Aerosol Sci. 2025;188:106595. https://doi.org/10.1016/j.jaerosci.2025.106595
[Google Scholar]

[19] Jo S., Kim T., Iyer V.G., Im W.. CHARMM-GUI: A web-based graphical user interface for CHARMM. J Comput Chem. 2008;29:1859-1865. https://doi.org/10.1002/jcc.20945
[Google Scholar]

[20] Johnston R.C., Yao K., Kaplan Z., Chelliah M., Leswing K., Seekins S., Watts S., Calkins D., Chief Elk J., Jerome S.V., Repasky M.P., Shelley J.C.. Epik: P Ka and protonation state prediction through machine learning. J Chem Theory Comput. 2023;19:2380-2388. https://doi.org/10.1021/acs.jctc.3c00044
[Google Scholar]

[21] Kamal A.M., Nabih N.A., Rakha N.M., Sanad E.F.. Upregulation of necroptosis markers RIPK3/MLKL and their crosstalk with autophagy-related protein Beclin-1 in primary immune thrombocytopenia. Clin Exp Med. 2023;23:447-456. https://doi.org/10.1007/s10238-022-00839-8
[Google Scholar]

[22] Kar S., Leszczynski J.. Open access in silico tools to predict the ADMET profiling of drug candidates. Expert Opin Drug Discov. 2020;15:1473-1487. https://doi.org/10.1080/17460441.2020.1798926
[Google Scholar]

[23] Kashiwagi H., Tomiyama Y.. Pathophysiology and management of primary immune thrombocytopenia. Int J Hematol. 2013;98:24-33. https://doi.org/10.1007/s12185-013-1370-4
[Google Scholar]

[24] Katsila T., Spyroulias G.A., Patrinos G.P., Matsoukas M.T.. Computational approaches in target identification and drug discovery. Comput Struct Biotechnol J. 2016;14:177-184. https://doi.org/10.1016/j.csbj.2016.04.004
[Google Scholar]

[25] Kim M.O., Nichols S.E., Wang Y., McCammon J.A.. Effects of histidine protonation and rotameric states on virtual screening of m. tuberculosis RmlC. J Comput Aided Mol Des. 2013;27:235-246. https://doi.org/10.1007/s10822-013-9643-9
[Google Scholar]

[26] Kumari R., Rathi R., Pathak S.R., Dalal V.. Computational investigation of potent inhibitors against YsxC: Structure-based pharmacophore modeling, molecular docking, molecular dynamics, and binding free energy. J Biomol Struct Dyn. 2023;41:930-941. https://doi.org/10.1080/07391102.2021.2015446
[Google Scholar]

[27] Lemak A.S., Balabaev N.K.. On the Berendsen thermostat. Mol Simul. 1994;13:177-187. https://doi.org/10.1080/08927029408021981
[Google Scholar]

[28] Li J., van der Wal D.E., Zhu G., Xu M., Yougbare I., Ma L., Vadasz B., Carrim N., Grozovsky R., Ruan M., Zhu L., Zeng Q., Tao L., Zhai Z.M., Peng J., Hou M., Leytin V., Freedman J., Hoffmeister K.M., Ni H.. Desialylation is a mechanism of Fc-independent platelet clearance and a therapeutic target in immune thrombocytopenia. Nat Commun. 2015;6:7737. https://doi.org/10.1038/ncomms8737
[Google Scholar]

[29] Li Q., Shah S.. Structure-based virtual screening. In: Protein Bioinformatics. New York, NY: Springer New York; 2017. p. :111-124. https://doi.org/10.1007/978-1-4939-6783-4_5
[Google Scholar]

[30] Liu S.‐Y, Zhang X.‐M., Sun R.‐J., Zhu J.‐J, Yuan D., Shan N.‐N. Abnormal expression of autophagy‐related proteins in immune thrombocytopenia. Scand J Immunol. 2021;93 https://doi.org/10.1111/sji.12992
[Google Scholar]

[31] Liu X.G., Hou Y., Hou M.. How we treat primary immune thrombocytopenia in adults. J Hematol Oncol. 2023;16:4. https://doi.org/10.1186/s13045-023-01401-z
[Google Scholar]

[32] Liu Z., Mei T.. Immune thrombocytopenia induces autophagy and suppresses apoptosis in megakaryocytes. Mol Med Rep. 2018;18:4016-4022. https://doi.org/10.3892/mmr.2018.9373
[Google Scholar]

[33] Maestro S.J.S.. Llc, New York, Ny. Maestro. 2020;2020:682.
[Google Scholar]

[34] Mali S.N., Chaudhari H.K.. Computational studies on Imidazo[1,2-a] Pyridine-3-carboxamide analogues as antimycobacterial agents: Common pharmacophore generation, atom-based 3D-QSAR, molecular dynamics simulation, QikProp, molecular docking and prime MMGBSA approaches. Pharmsci. 2018;5:12-23. https://doi.org/10.2174/1874844901805010012
[Google Scholar]

[35] Martínez L.. Automatic identification of mobile and rigid substructures in molecular dynamics simulations and fractional structural fluctuation analysis. PLoS One. 2015;10:e0119264. https://doi.org/10.1371/journal.pone.0119264
[Google Scholar]

[36] Martonák R., Laio A., Parrinello M.. Predicting crystal structures: The Parrinello-Rahman method revisited. Phys Rev Lett. 2003;90:075503. https://doi.org/10.1103/PhysRevLett.90.075503
[Google Scholar]

[37] Moumbock A.F.A., Li J., Tran H.T.T., Hinkelmann R., Lamy E., Jessen H.J., Günther S.. ePharmaLib: A versatile library of e-Pharmacophores to address small-molecule (poly-)pharmacology. J Chem Inf Model. 2021;61:3659-3666. https://doi.org/10.1021/acs.jcim.1c00135
[Google Scholar]

[38] Pathak R.K., Singh D.B., Sagar M., Baunthiyal M., Kumar A.. Computational approaches in drug discovery and design. In: Computer-Aided Drug Design. Singapore: Springer; 2020. p. :1-21. https://doi.org/10.1007/978-981-15-6815-2_1
[Google Scholar]

[39] Provan D., Newland A.C.. Current management of primary immune thrombocytopenia. Adv Ther. 2015;32:875-887. https://doi.org/10.1007/s12325-015-0251-z
[Google Scholar]

[40] Provan D., Semple J.W.. Recent advances in the mechanisms and treatment of immune thrombocytopenia. EBioMedicine. 2022;76:103820. https://doi.org/10.1016/j.ebiom.2022.103820
[Google Scholar]

[41] Sadybekov A.V., Katritch V.. Computational approaches streamlining drug discovery. Nature. 2023;616:673-685. https://doi.org/10.1038/s41586-023-05905-z
[Google Scholar]

[42] Sandal R., Mishra K., Jandial A., Sahu K.K., Siddiqui A.D.. Update on diagnosis and treatment of immune thrombocytopenia. Expert Rev Clin Pharmacol. 2021;14:553-568. https://doi.org/10.1080/17512433.2021.1903315
[Google Scholar]

[43] Sargsyan K., Grauffel C., Lim C.. How molecular size impacts RMSD applications in molecular dynamics simulations. J Chem Theory Comput. 2017;13:1518-1524. https://doi.org/10.1021/acs.jctc.7b00028
[Google Scholar]

[44] Schrödinger L.J.S.S.. Version 2, 2017-2011. New York, NY: Schrödinger, LLC; 2017.

[45] Shukla R., Tripathi T.. Molecular dynamics simulation in drug discovery: opportunities and challenges. In: Innovations and Implementations of Computer Aided Drug Discovery Strategies in Rational Drug Design. Singapore: Springer Singapore; 2021. p. :295-316. https://doi.org/10.1007/978-981-15-8936-2_12
[Google Scholar]

[46] Thillainayagam M., Malathi K., Ramaiah S.. In-silico molecular docking and simulation studies on novel chalcone and flavone hybrid derivatives with 1, 2, 3-triazole linkage as vital inhibitors of Plasmodium falciparum dihydroorotate dehydrogenase. J Biomol Struct Dyn. 2018;36:3993-4009. https://doi.org/10.1080/07391102.2017.1404935
[Google Scholar]

[47] Van Der Spoel D., Lindahl E., Hess B., Groenhof G., Mark A.E., Berendsen H.J.. GROMACS: Fast, flexible, and free. J Comput Chem. 2005;26:1701-1718. https://doi.org/10.1002/jcc.20291
[Google Scholar]

[48] Vanommeslaeghe K., MacKerell A.D.. Automation of the CHARMM general force field (CGenFF) i: Bond perception and atom typing. J Chem Inf Model. 2012;52:3144-3154. https://doi.org/10.1021/ci300363c
[Google Scholar]

[49] Wang C.Y., Ma S., Bi S.J., Su L., Huang S.Y., Miao J.Y., Ma C.H., Gao C.J., Hou M., Peng J.. Enhancing autophagy protects platelets in immune thrombocytopenia patients. Ann Transl Med. 2019;7:134. https://doi.org/10.21037/atm.2019.03.04
[Google Scholar]

[50] Wang Z., Sun H., Shen C., Hu X., Gao J., Li D., Cao D., Hou T.. Combined strategies in structure-based virtual screening. Phys Chem Chem Phys. 2020;22:3149-3159. https://doi.org/10.1039/c9cp06303j
[Google Scholar]

[51] Xu H.D., Qin Z.H.. Beclin 1, Bcl-2 and Autophagy. Adv Exp Med Biol. 2019;1206:109-126. https://doi.org/10.1007/978-981-15-0602-4_5
[Google Scholar]

[52] Yu L., Chen Y., Tooze S.A.. Autophagy pathway: Cellular and molecular mechanisms. Autophagy. 2018;14:207-215. https://doi.org/10.1080/15548627.2017.1378838
[Google Scholar]

[53] Zhou H., Hou Y., Liu X., Qiu J., Feng Q., Wang Y., Zhang X., Min Y., Shao L., Liu X., Li G., Li L., Yang L., Xu S., Ni H., Peng J., Hou M.. Low-dose decitabine promotes megakaryocyte maturation and platelet production in healthy controls and immune thrombocytopenia. Thromb Haemost. 2015;113:1021-1034. https://doi.org/10.1160/TH14-04-0342
[Google Scholar]

[54] Zufferey A., Kapur R., Semple J.W.. Pathogenesis and therapeutic mechanisms in immune thrombocytopenia (ITP) J Clin Med. 2017;6:16. https://doi.org/10.3390/jcm6020016
[Google Scholar]

Sr.	Catalog ID	Glide score (kcal/mol)	Hydrogen bonds	Hydrophobic interactions
1	Z1672440949	-5.766	GLU144, ARG155	LEU1, LEU3, CYS151, LEU152, PRO153, TYR154
2	Z9255349517	-5.501	ARG149, AR155, GLU144	LEU3, LEU1, CYS151, PHE150
3	Z9255355511	-5.382	LEU1, LEU3, GLU144	CYS151, LEU152, PRO153, TYR154
4	Z7837265532	-5.337	ARG155, LEU1	LEU3, CYS151, LEU152, PRO153, TYR154
5	Z9255356311	-5.407	LEU1, GLU144, LYS140, ARG155	LEU3, CYS151, LEU152, PRO153, TYR514
6	Z9255355469	-5.191	GLU144, ARG155	LEU1, LEU3, CYS151, LEU152, PRO153, TYR154
7	Z9255355525	-5.127	LEU1, PHE150, ARG149, GLU144	CYS151
8	Z9255355511	-5.103	LEU1, LEU3, GLU144, LYS140	PHE150, CYS151, LEU152, PRO153, TYR154

Identification of Beclin-1 inhibitors to treat Immune thrombocytopenia by structure-based virtual screening and molecular modelling approaches

Abstract

Keywords

Autophagy-related proteins

Beclin-1

Immune thrombocytopenia (ITP)

Molecular docking

MD simulation

Virtual screening

1. Introduction

2. Materials and Methods

2.1 Generation of pharmacophore model

2.2 Virtual screening

2.3 Molecular docking

2.4 Molecular interaction analysis

2.5 Drug likeness and toxicity analysis

2.6 MD simulation

3. Results

3.1 Pharmacophore modeling

3.2 Virtual screening

3.3 Molecular docking

3.4 Post docking analysis

3.5 ADMET analysis

3.6 Binding pose alignment

3.7 MD simulation

3.7.1 RMSD

3.7.2 RMSF

3.7.3 Radius of gyration

3.7.4 Hydrogen bonding

4. Discussion

5. Conclusions

CRediT authorship contribution statement

Declaration of competing interest

Declaration of Generative AI and AI-assisted technologies in the writing process

References

Suggested read for related articles: