|Year : 2018 | Volume
| Issue : 1 | Page : 18-23
In silico identification of potent small molecule inhibitors targeting epidermal growth factor receptor 1
Zheng Shi1, Jie Chen2, Xiaolan Guo1, Lijia Cheng1, Xiaoheng Guo1, Tian Yu1
1 School of Medicine, Sichuan Industrial Institute of Antibiotics, Chengdu University, Chengdu, China
2 Central Laboratory of Clinical Medicine, Sichuan Academy of Medical Science and Sichuan Provincial People's Hospital, Chengdu, China
|Date of Web Publication||8-Mar-2018|
Dr. Lijia Cheng
School of Medicine, Sichuan Industrial Institute of Antibiotics, Chengdu University, Chengdu
Dr. Xiaoheng Guo
School of Medicine, Sichuan Industrial Institute of Antibiotics, Chengdu University, Chengdu
Source of Support: None, Conflict of Interest: None
Background: The receptor tyrosine kinase of the epidermal growth factor receptor (EGFR, ErbB) family played an important role in multisignaling pathways, which controlled numerous biological activities including proliferation, differentiation, apoptosis, etc. EGFR abnormalities have been associated with a variety of human tumors, which was a well-characterized target for cancer treatment. It was known to all that drug repositioning has been considered as a useful tool to accelerate the process of drug development.
Materials and Methods: Herein, a total of 1408 small molecule drugs approved by the Food and Drug Administration (FDA) were employed to identify potential EGFR inhibitors by a series of bioinformatics approaches, including virtual screening and molecular dynamics (MD) simulations.
Results: According to the docking score, five small molecules were chosed for further MD simulations. Following the 5 ns MD simulations, ZINC03830276 (Benzonatate) were finally recognized as “new use” of FDA-approved EGFR-targeting drug.
Conclusions: Our findings suggested that the small molecule ZINC03830276 (Benzonatate) could be a promising EGFR inhibitor candidate and may also provide new ideas for designing more potent EGFR inhibitors for the future study.
Keywords: Drug discovery, drug repositioning, epidermal growth factor receptor, kinase inhibitor, virtual screening
|How to cite this article:|
Shi Z, Chen J, Guo X, Cheng L, Guo X, Yu T. In silico identification of potent small molecule inhibitors targeting epidermal growth factor receptor 1. J Can Res Ther 2018;14:18-23
|How to cite this URL:|
Shi Z, Chen J, Guo X, Cheng L, Guo X, Yu T. In silico identification of potent small molecule inhibitors targeting epidermal growth factor receptor 1. J Can Res Ther [serial online] 2018 [cited 2021 Jun 24];14:18-23. Available from: https://www.cancerjournal.net/text.asp?2018/14/1/18/226739
| > Introduction|| |
Epidermal growth factor receptor (EGFR) is a transmembrane receptor tyrosine that belongs to the ErbB superfamily. There were four-related conserved structural members, including EGFR, ErbB-2, ErbB-3, and ErbB-4 (also known as HER-2, HER-3 and HER-4) in this family. The ErbB superfamily members were activated by binding with its specific ligands. Once the EGF bound to the ligand binding domain, the EGFR formed a homodimer or heterodimer with other EGFR family members. Subsequently, the intrinsic kinase domain of EGFR was activated, resulting in autophosphorylation on specific tyrosine residues within the cytoplasmic tail. Then, a number of signal transduction pathway cascades would be initiated with the downstream effector, eventually leading to DNA synthesis and cell proliferation.
EGFR played a significant role in the regulation of several signaling cascades, for example, PI3K/AKT/mTOR and RAS/RAF/MEK, which ultimately induced cell proliferation, survival, etc. Overexpression and mutational activation of EGFR were associated with the development and progression of multiple aggressive human cancers. Therefore, negative regulation of EGFR has been clinically proposed as a therapeutic approach for the treatment of various cancer types.
There was an ATP binding site in EGFR for binding ATP for autophosphorylation in the intracellular tyrosine kinase domain. The ATP binding site was considered to be an effective site binding with inhibitors to inhibit the receptor by blocking the autophosphorylation of the tyrosine residue. Although a variety of potent EGFR inhibitors has been continuously expanding, the discovery of novel small molecule inhibitors has been still a hot spot. Virtually screening has become a powerful method for searching novel lead compounds in drug development. Hitherto, plenty of small molecule kinase inhibitors were successfully recognized, such as PLK1 inhibitors, CDKs inhibitors, SRC inhibitors.
The development of novel drugs has been expensive and time-consuming. Discovering new pharmacological activity from existing drugs, or drug repositioning, allowed for dramatic acceleration of new drug discovery and development. To our the best of our knowledge, a number of known drugs have new use. For instance, Tretinoin, a standby for the treatment of severe acne, has been currently applied for the treating acute promyelocytic leukemia. Thalidomide, originally used as a hypnotic and now has been applied for the treatment of multiple myeloma. Hitherto, at least 17 existing drugs were in various stages of clinical and animal testing for new uses.
In this study, we aimed to apply molecular docking-based virtual screening to filter 1408 Food and Drug Administration (FDA)-approved small molecule drugs, to identify “new use” anti-EGFR inhibitors. A series of bioinformatics methods was proposed to identify “uses for old drugs” as potential EGFR inhibitors. First, five potential drugs were chosed based on their amber score. Furthermore, molecular dynamics (MD) simulations were utilized to recognize the affinities and stabilities of EGFR with abovementioned five drugs. Finally, based on the above analysis, ZINC03830276 (Benzonatate) were successfully screened, which exerted potential inhibitory effects toward EGFR protein.
| > Materials and Methods|| |
The initial three-dimensional (3D) of the X-ray crystal structure of EGFR kinase domain (Protein Database Bank [PDB]: 5HG8) and its original ligand (634) were downloaded from the PDB (http://www.rcsb.org/pdb/home/home.do). In addition, the EGFR and the candidate 1408 small molecule agents were prepared with USFC chimera (version 1.8). The preparation process was further managed, including hydrogen atom and standard charge additions, as well as the elimination of solvent molecule. Subsequently, FDA-approved small molecule agents were collected for further screening (http://zinc.docking.org/catalogs/specsnp). All these molecules were in ready-to-dock 3D format structures and commercially available.
USFC DOCK (6.4 program) was a typical program which has been utilized to screen in massive compound libraries toward a target for identifying potential drug leads. The molecular docking calculations were performed using UCSF DOCK. Amber force-field parameters and flexible ligand docking method were applied to predict inhibitors targeting EGFR kinase domain. For improving the accuracy, the binding site was selected within 10.0 Š root mean square deviations (RMSD) of original ligand.
In addition, flexible-ligand docking to a rigid receptor with grid-based scoring was carried out, in which the small molecule drugs could be structurally rearranged in response to the EGFR. The docking pose was first ranked which based on the nonbonded terms of the molecular mechanic force field. Moreover, the top-ranked 100 grid-generated hits were reranked by Amber score function. The ranked structures of these molecules were further redocked into the kinase domain of EGFR by Hawkins GB/SA Scoring, Amber scoring, and descriptor scoring algorithm, respectively. Hawkins GB/SA Score was a pairwise descreening approximation which applied in the calculations of electrostatic energy contribution. It has also been developed to account for the low-dielectric region that might form between the ligand and receptor during docking processes. Amber score enabled both the compound and receptor as flexible to reproduce the so-called “induced fit,” and it took solvation energies into consideration. After amber scores and docking pose, 5 EGFR-approved drug complexes were eventually selected for further MD simulations analysis.
Molecular dynamics simulations
The GROMACS (version 4.5, http://www.gromacs.org/) package was involved to perform MD simulations analysis of the dynamic behavior of EGFR complexes with AMBER99SB force field, and the receptor topology was constructed by pdb2gmx. Furthermore, the ligand topology was generated through the PRODRG2 server. Moreover, the unit cell as cubic box and filled it with single-point charge water molecules was defined in 1.0 solute-wall minimum distance. After a steepest descent energy minimization with particle mesh Ewald  at every step, EGFR-ligand complexes were equilibrated 100 ps under constant volume (NVT) and under constant pressure (NPT) condition for 1 bar, respectively. In addition, position restraints were simultaneously applied to EGFR kinase domain and its ligands. On the completion of two equilibration phases, production MD simulations were conducted for 5 ns after taking away the position restraints. During data collection phase, the temperature and pressure were maintained with the same methods as in NVT and NPT phase, respectively. Moreover, the EGFR-drugs complexes' coordinates were saved every 0.2 ps. The results of all structural analysis were processed by modules in GROMACS packages.
Subsequently, LIGPLOT+ software (Version 1.4.5) was taken advantage to discover the hydrogen bonds and hydrophobic interactions between key residues of EGFR and 1408 FDA small molecules. LIGPLOT+ was commonly applied to plot protein-ligand (LIGPLOT) and protein-protein (DIMPLOT) interactions. The 3D protein-ligand complexes generated during MD simulations were converted into 2D diagrams using the LIGPLOT algorithm.
| > Results|| |
Molecular docking for potential epidermal growth factor receptor inhibitors
To uncover the mechanisms of ligand-protein interactions, 1408 small molecule drugs were docked into the kinase domain of EGFR. In general, the lower the docking score of ligands was, the higher the affinity binding to receptor. Consequently, we finally achieved top10 drugs based on their amber score, and the amber scores of the 10 ligands in complex with EGFR were ranged from −36.4044 to −50.8367. The molecular docking results of EGFR in complex with top10 drugs were shown [Table 1]. Subsequently, the top 5 ranked compounds were selected for the following MD simulations analysis, including ZINC49583080 (Dofetilide), ZINC00537804 (Glisoxepid), ZINC03830276 (Benzonatate), ZINC53230301 (Cabergoline), and ZINC33956087 (Iloprost). All the grid score and top10 amber score were shown [Table S1 [Additional file 1]].
|Table 1: Molecular docking results of epidermal growth factor receptor in complex with Food and Drug Administration-approved small molecule drugs|
Click here to view
Molecular dynamics simulations
MD simulations were performed to assess the dynamic interactions between receptors and ligands, which could indicate the time-evolving structure ensemble, represent the solution conformation of the proteins, exhibit different dynamic processes according to different ligands. The Ecoul, potential energy, kinetic energy, and total energy were calculated with the g_energy program available in Gromacs, extracting information from “.edr” files were generated during the MD simulations. The interaction energy (the sum of electrostatic and van der Waals interactions) between the EGFR and its ligands was subsequently considered [Figure 1] and [Table 2].
|Figure 1: Root mean square deviations of epidermal growth factor receptor-drug complexes backbone atoms during the molecular dynamics simulations|
Click here to view
|Table 2: Molecular dynamics simulations results of epidermal growth factor receptor in complex with Food and Drug Administration-approved small molecule drugs|
Click here to view
RMSD from the initial confirmation were a dominating criterion which was applied to access the stability of complexes. RMSD of the atomic positions to examine the conformational variations of EGFR complexes for 5 ns were developed to evaluate the dynamic stability of each among 5 ligands. Herein, the overall conformational evolution of the EGFR-ligand complexes during the room-temperature simulation was subsequently analyzed. As the obtained results [Figure 1], EGFR-ZINC00537084 (Glisoxepid) and EGFR-ZINC49583080 (Dofetilide) showed a similarly sharp rise during the first 0.5 ns. It was worth noting that EGFR-ZINC00537804 (Glisoxepid) system could reach its stability after 3.5 ns, and the EGFR-ZINC03830276 (Benzonatate) system showed the trend of rise within 3 ns while the EGFR-ZINC49583080 (Dofetilide) system exhibited maximum deviation. In general, we could infer that the EGFR-ZINC03830276 (Benzonatate) system was much more stable than the others, which suggested this compound might exert a stronger affinity with EGFR [Figure 1] and [Table 2].
Root-mean-square fluctuation (RMSF) could compare the flexibility of each residue in receptor-ligand complexes. In this work, a detailed analysis of RMSF versus the residue number in the complexes was illustrated [Figure 2]. EGFR-ZINC49583080 (Dofetilide) complex exhibited higher flexibility while EGFR-ZINC03830276 (Benzonatate) complex showed lower flexibility in EGFR-ligand complexes.
|Figure 2: Root-mean-square fluctuation of the backbone atoms versus the residue number of the epidermal growth factor receptor-drug complexes during molecular dynamics simulations|
Click here to view
At the same time, the secondary structure could describe the stability of the complexes. From [Figure 3], it observed that the residues 160–170 of EGFR-ZINC53230301 (Cabergoline) changed from A-helix to turn while the other systems remained relatively stable. Throughout the whole dynamic simulation processes, no water molecules were observed within the hydrophobic cavity.
|Figure 3: Plots for the composition of interaction energy of epidermal growth factor receptor with different small molecules during the molecular modeling|
Click here to view
A particular comparison of the binding fashion might provide a novel insight into the key residues which formed hydrogen bonds and hydrophobic interactions with EGFR to stabilize the corresponding inhibitors. Herein, the key residues were plotted with LIGPLOT + to analyze the mechanism of the protein-ligand interaction. We have analyzed EGFR with its original ligand (634) and ZINC03830276 (Benzonatate) [Figure 4]. The interactions between the EGFR residues and the corresponding inhibitor for each binding model could be observed [Figure 4]. For example, in the EGFR-ZINC03830276 (Benzonatate) system, hydrogen bonds were formed between the residues Lys716 and Lys728 with EGFR.
|Figure 4: Bind fashion comparison between epidermal growth factor receptor and ZINC03830276 (Benzonatate) complexes and with its original ligand (634). Hydrophobic interactions were represented with dark cyan arcs. Hydrogen bonds were indicated with broken black lines. Ligands were represented in purple. (For interpretation of the references to color in this figure legend, the reader could refer to the web version of this article)|
Click here to view
| > Discussion|| |
EGFR family regulated numerous biological functions including cell proliferation, differentiation, survival and apoptosis. EGFR and its downstream signaling pathways have been one of the most attractive pathways in terms of targeting anticancer drug development. The dysregulated expression of EGFR was implicated in a range of malignancies (breast, ovarian, colon, prostate, etc.), thus holding appeal as an effective therapeutic target for cancer drug discovery. Currently, numerous compounds designed as EGFR inhibitors were still under evaluation in preclinical and clinical trials for the treatment of cancer.
Currently, there were two types of EGFR inhibitors which have been applied in clinic trails. First, monoclonal antibodies could target the extracellular domain, such as cetuximab and panitumumab; the other one was small molecule inhibitors, which could target the intracellular tyrosine kinase domain, including erlotinib, lapatinib, and afatinib.,,,
Of note, discovery of new effective EGFR inhibitors from natural product database has been usually time and humanpower consuming. Therefore, to dramatically accelerate anticancer drug discovery, repurposing the existing drugs for previously unknown therapeutic usages may serve as an excellent complementary strategy. Old drugs can quickly enter clinical trials for newly recognized indications with existing drug dosage regimen. Therefore, discovery of old drugs for new use has been a powerful and effective approach for cancer therapy. In addition, all the bioinformatics methods mentioned in our work now have been the most popular and promising approaches in drug discovery.
It is known to all that EGFR mutation related to tyrosine kinase inhibitors responsiveness in cancer has become an important issue for EGFR-harbor cancer patients. The first-generation ATP-competitive EGFR inhibitors were the first-line treatment of EGFR-mutated cancer. An acquired secondary mutation (T790M) was noticed shortly after the treatment. Clinical studies indicated that such mutation altered the binding site and made first-generation inhibitors ineffective. Second-generation inhibitors such as afatinib exerted inhibitory activity for both wtEGFR and EGFR T790M while the third-generation inhibitors such as rociletinib inhibited only T790M resistance mutant and spared the wild type.
In the current study, virtual screening was adopted to recognize the potential EGFR inhibitors among the known FDA approved drugs. In a word, according to the RMSD and RMSF of MD simulations, we considered that ZINC03830276 (Benzonatate) was successfully selected to exert potential inhibitory effects toward EGFR protein. More importantly, we further analyze the binding fashion between EGFR-ZINC03830276 (Benzonatate) and EGFR-634 with LigPlot +. The results indicated that hydrogen bonds were formed between the residues Lys716 and Lys728 with EGFR.
The analysis of crystal structure of ZINC03830276 (Benzonatate)-EGFR complexes referred the competition for the ATP-binding pocket, and thus inhibiting further phosphorylation of EGFR kinase and decreasing its kinase activity, eventually inducing cancer cell death [Figure 5]. Furthermore, the binding mechanisms of these three most potential inhibitors were further analyzed with LigPlot +. Ligplot + diagrams could demonstrate the hydrophobic and H-bond interactions between EGFR and the 4 most potential inhibitors. Hydrogen bonds and hydrophobic interactions between the residues and the ligand were represented by dotted lines and arcs, respectively. Blue circles and ellipses highlighted the key residues identified from the balance time of MD simulations. In general, the more energy the amino acid residues could contribute, the higher the ligand binding affinity was. Finally, Lys716 and Pro794 were identified as the key residues based on the MD simulations.
|Figure 5: Potential small molecule drugs targeting epidermal growth factor receptor|
Click here to view
| > Conclusions|| |
In this study, we first assemble a library of FDA approved clinical drugs. Second, 10 old drugs were recognized to bind EGFR based on their docking score. Moreover, top five small molecules were manually selected for further MD simulation analysis. A small molecule, namely, ZINC03830276 (Benzonatate) exerting relatively stable affinity was successfully chosen according to their MD analysis. The most potential EGFR inhibitor bind mechanism was further analyzed by LigPlot +. The small molecule compound could potentially compete with ATP binding at the cytoplasmic catalytic kinase domain of EGFR, thus blocking the autophosphorylation and activation of EGFR. Then, the downstream proliferation and survival signals would be inhibited, leading to increased apoptosis and decreased cellular proliferation. Herein, we eventually identified approved small molecule drugs, ZINC03830276 (Benzonatate) as the potential inhibitors targeting EGFR, which could effectively treat the EGFR harbored cancer patients. In conclusion, our findings may not only provide an efficient and cost-effective virtual screening approach to identify potent EGFR inhibitors but also pave the new road for the repositioning of existing approved drugs for the further development of EGFR inhibitors.
We were grateful to Miss Wen-wen Li (University College London), Rong Sun (Sichuan University) for providing constructive suggestions.
Financial support and sponsorship
This work was supported in part by Scientific and Technological Funds for Young Scientists of Sichuan (2017JQ0060), the National Natural Science Foundation of China (No. 51402027) and the science and technology support program of Science and Technology Department of Sichuan (2016NZ0060, 2017NZ0046).
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Choowongkomon K, Sawatdichaikul O, Songtawee N, Limtrakul J. Receptor-based virtual screening of EGFR kinase inhibitors from the NCI diversity database. Molecules 2010;15:4041-54.
Li S, Sun X, Zhao H, Tang Y, Lan M. Discovery of novel EGFR tyrosine kinase inhibitors by structure-based virtual screening. Bioorg Med Chem Lett 2012;22:4004-9.
Sawatdichaikul O, Hannongbua S, Sangma C, Wolschann P, Choowongkomon K. In silico
screening of epidermal growth factor receptor (EGFR) in the tyrosine kinase domain through a medicinal plant compound database. J Mol Model 2012;18:1241-54.
Gupta AK, Bhunia SS, Balaramnavar VM, Saxena AK. Pharmacophore modelling, molecular docking and virtual screening for EGFR (HER 1) tyrosine kinase inhibitors. SAR QSAR Environ Res 2011;22:239-63.
Shi Z, An N, Lu BM, Zhou N, Yang SL, Zhang B, et al.
Identification of novel kinase inhibitors by targeting a kinase-related apoptotic protein-protein interaction network in HeLa cells. Cell Prolif 2014;47:219-30.
Wang ZJ, Wan ZN, Chen XD, Wu CF, Gao GL, Liu R, et al. In silico
identification of novel kinase inhibitors by targeting B-raf (v660e) from natural products database. J Mol Model 2015;21:102.
Hoelder S, Clarke PA, Workman P. Discovery of small molecule cancer drugs: Successes, challenges and opportunities. Mol Oncol 2012;6:155-76.
Chong CR, Sullivan DJ Jr. New uses for old drugs. Nature 2007;448:645-6.
Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H. et al.
The protein data bank/nucleic acids research. J Nucleic Acids Res 1989;17:3588.
Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, et al.
UCSF Chimera – A visualization system for exploratory research and analysis. J Comput Chem 2004;25:1605-12.
Irwin JJ, Shoichet BK. ZINC – A free database of commercially available compounds for virtual screening. J Chem Inf Model 2005;45:177-82.
Lang PT, Brozell SR, Mukherjee S, Pettersen EF, Meng EC, Thomas V, et al.
DOCK 6: Combining techniques to model RNA-small molecule complexes. RNA 2009;15:1219-30.
Shi Z, Yu T, Sun R, Wang S, Chen XQ, Cheng LJ, et al.
Discovery of novel human epidermal growth factor receptor-2 inhibitors by structure-based virtual screening. Pharmacogn Mag 2016;12:139-44.
Dolinsky TJ, Nielsen JE, McCammon JA, Baker NA. PDB2PQR: An automated pipeline for the setup of Poisson-Boltzmann electrostatics calculations. Nucleic Acids Res 2004;32:W665-7.
Shi Z, Wang ZJ, Xu HL, Tian Y, Li X, Bao JK, et al.
Modeling, docking and dynamics simulations of a non-specific lipid transfer protein from Peganum harmala
L. Comput Biol Chem 2013;47:56-65.
Pronk S, Páll S, Schulz R, Larsson P, Bjelkmar P, Apostolov R, et al.
GROMACS 4.5: A high-throughput and highly parallel open source molecular simulation toolkit. Bioinformatics 2013;29:845-54.
Morris GM, Lim-Wilby M. Molecular docking. Methods Mol Biol 2008;443:365-82.
Van Der Spoel D, Lindahl E, Hess B, Groenhof G, Mark AE, Berendsen HJ, et al.
GROMACS: Fast, flexible, and free. J Comput Chem 2005;26:1701-18.
Brozell SR, Mukherjee S, Balius TE, Roe DR, Case DA, Rizzo RC, et al.
Evaluation of DOCK 6 as a pose generation and database enrichment tool. J Comput Aided Mol Des 2012;26:749-73.
Songtawee N, Gleeson MP, Choowongkomon K. Computational study of EGFR inhibition: Molecular dynamics studies on the active and inactive protein conformations. J Mol Model 2013;19:497-509.
Sun R, Li X, Li Y, Zhang X, Li X, Li X, et al.
Screening of novel inhibitors targeting lactate dehydrogenase A via four molecular docking strategies and dynamics simulations. J Mol Model 2015;21:133.
Laskowski RA, Swindells MB. LigPlot+: Multiple ligand-protein interaction diagrams for drug discovery. J Chem Inf Model 2011;51:2778-86.
Wallace AC, Laskowski RA, Thornton JM. LIGPLOT: A program to generate schematic diagrams of protein-ligand interactions. Protein Eng 1995;8:127-34.
Jänne PA, Yang JC, Kim DW, Planchard D, Ohe Y, Ramalingam SS, et al.
AZD9291 in EGFR inhibitor-resistant non-small-cell lung cancer. N
Engl J Med 2015;372:1689-99.
Vaishnavi A, Le AT, Doebele RC. TRKing down an old oncogene in a new era of targeted therapy. Cancer Discov 2015;5:25-34.
Yang SC, Chang SS, Chen HY, Chen CY. Identification of potent EGFR inhibitors from TCM database@Taiwan. PLoS Comput Biol 2011;7:e1002189.
Puvanenthiran S, Essapen S, Seddon AM, Modjtahedi H. Impact of the putative cancer stem cell markers and growth factor receptor expression on the sensitivity of ovarian cancer cells to treatment with various forms of small molecule tyrosine kinase inhibitors and cytotoxic drugs. Int J Oncol 2016;49:1825-38.
Gudala S, Khan U, Kanungo N, Bandaru S, Hussain T, Parihar M, et al.
Identification and pharmacological analysis of high efficacy small molecule inhibitors of EGF-EGFR interactions in clinical treatment of non-small cell lung carcinoma: A Computational approach. Asian Pac J Cancer Prev 2015;16:8191-6.
Zibelman M, Mehra R. Overview of current treatment options and investigational targeted therapies for locally advanced squamous cell carcinoma of the head and neck. Am J Clin Oncol 2016;39:396-406.
Asegaonkar SB, Asegaonkar BN, Takalkar UV, Advani S, Thorat AP. C-reactive protein and breast cancer: New insights from old molecule. Int J Breast Cancer 2015;2015:145647.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2]