Identification of potential binding pocket on viral oncoprotein HPV16 E6: a promising anti-cancer target for small molecule drug discovery
- 347 Downloads
Several human cancers, especially cervical cancer are caused by the infection of high risk strains of human papillomaviruses (HPV), notably HPV16. It is implicated that the oncoprotein E6 expressed from HPV, is inhibiting the apoptotic pathway by binding to adaptor molecule FADD (Fas-associated death domain). Inhibiting E6 interactions with FADD could provide a promising treatment for cervical cancer. There are few small molecules reported to inhibit such interactions. However, the FADD binding site information on the HPV E6 is not currently available. This binding site information may provide an opportunity to design new small molecule inhibitors to treat E6 mediated cancers. In this study we report the possible binding pocket on HPV16 E6 oncoprotein by using activity data of reported inhibitors through a stepwise molecular modeling approach.
Blind docking and removing duplicates followed by visual inspection to determine ligand-receptor interactions provided 68 possible binding sites on the E6 protein. Individual docking of all known inhibitors lead to the identification of 28 pockets having some kind of correlation with their activity data. It was also observed that several of these pockets overlapped with each other, having some amino acids in common. Amino acids Leu50 and Cys51 were identified as key E6 residues for high affinity ligand binding which are seen in most of these pockets. In most cases, ligands demonstrated a hydrogen bond interaction with Cys51. Ala61, Arg131 and Gln107 were also frequently observed showing interactions among these pockets. A few amino acids unique to each ligand were also identified representing additional interactions at the receptor site.
After determining receptor-ligand interactions between E6 oncoprotein and the six known inhibitors, the amino acids Cys51, Leu50, Arg102, Arg131, Leu67, Val62, and Gln107 were identified to have importance in E6 inhibition. It was generally observed that Leu50 and Cys51 are necessary for high binding affinity with Cys51 being essential for hydrogen bonding. This study identified a potential binding pocket for the E6 inhibitors. Identification of the ligand binding pocket helps to design novel inhibitors of HPV16 E6 oncoprotein as a promising treatment for cervical cancer.
KeywordsHuman papilloma virus HPV E6 Myricetin Flavonol Molecular modelling
Fas-associated death domain
First apoptosis signal;
Tumor necrosis factor
TNF receptor associated death domain protein
TNF-related apoptosis-inducing ligand
Human papillomavirus (HPV) infection is one of the major causes of cervical cancer in women worldwide [1, 2, 3]. To date, more than 200 different HPVs have been characterized, and new types are regularly being added. These viruses are sub grouped into mucosal and cutaneous HPVs according to their ability to infect the mucosa or the skin of genital or upper respiratory tracts. Within each of these HPV groups, they are further sub-classified based on their risk to cause malignancy . Of all, HPV16 is known to be the most oncogenic type within the high-risk group [5, 6]. Prophylactic HPV vaccines are currently available, which help control or prevent certain types of cancers including cervical cancers. However, these vaccines are very expensive and have no utility in already infected patients [7, 8].
A few small molecules inhibitors were developed via high throughput screening for HPV E6 and E6AP interaction, with only modest activities [7, 19]. A library of flavonols were reported as inhibitors of HPV16 E6 and FADD and procaspase 8 . However, the specific binding site on the E6 protein is not known. Having the binding site information could provide insight for the rational design of improved inhibitors suitable for the treatment of HPV-mediated cancers. The development of small molecule inhibitors will provide a major advancement in the treatment of cervical cancers due to HPV. Over 99% of the patients having cervical lesions are positively identified with HPV DNA . These could be better alternatives or treatment options for women in countries that do not have access to or cannot afford vaccines. In this study, we identified the ligand binding pocket on the HPV16 E6 oncoprotein by using activity data of reported inhibitors through a systematic molecular modeling study. HPV16 E6 inhibitors could make a promising prophylactic treatment to prevent cancers in HPV infected patients.
Step 1: blind docking
In order to identify all pockets on the HPV16 E6 protein, blind docking was performed using a web-based program called SwissDock. The ligands myricetin, morin, galangin, kaempferol, 6-hydroxy flavonol, and quercetin (Fig. 2) were docked on chain ‘C’ from the unit cell of the crystal structure (PDB: 4GIZ). SwissDock generates all possible binding modes for each ligand and the most favorable binding modes at a given pocket were clustered. All ligand clusters were saved in an output file called “predictions file”. The predictions file provided Cluster Rank/Element Full Fitness and estimated binding free energy ΔG. Sample SwissDock predictions file data for the ligand morin docked in HPV16 E6, is shown in the Additional file 1: Figure S2, as viewed in Jmol applet within the SwissDock program. A cluster is a predicted binding pocket on the target protein, and the cluster rank/element represents the different conformations of the ligand in a certain cluster. Individual clusters and clusters with noticeably different conformations when viewed in SwissDock were saved separately from output predictions file for visualization. Overall, we identified 345 individual clusters after blind docking all six ligands.
Step 2: cluster visualization to determining receptor-ligand interactions
Discovery Studio visualizer was used to visualize the receptor ligand interactions for all individual clusters obtained from Step 1. Each cluster for every ligand was inspected for amino acids interacting with the ligand, types of bonds formed, the specific atoms involved and the distance between them. All of the interacting amino acids with the target receptor were noted for each cluster.
Step 3: removing duplicate clusters based on interacting amino acids
Step 4: docking into individual binding pockets
A schematic representation showing a process from GOLD Docking to predicting the final binding pocket is illustrated in Fig. 6. Binding pockets 1–22 are high binding pockets (HBP) with GOLD Scores above 40 for all ligands and binding pockets 23–28 have scores that correlate with the IC50 values of the ligands (ICBP).
Docking scores of flavonols at potential BPs along with experimental Ki values
Analysis of ligand-interacting amino acid residues of E6 oncoprotein
E6 flavonol binding pocket
In addition to the flavonol inhibitors of the HPV16 E6 interaction with procaspase 8 on which this study is based [20, 22], there have been other studies examining small molecule inhibitors for E6 interactions, and some other molecules have been proposed as a result of virtual screening [23, 24, 25, 26], discovered through screening [27, 28, 29, 30], or created by design [19, 31]. It is important to emphasize that each of these reports used assays measuring the interaction of the E6 protein with different protein partners and not with procaspase 8.
The earliest report of a small molecule inhibitor of the interaction between HPV16 E6 and the p53 tumor suppressor protein, was in 2005 for the flavone jaceosidin at a concentration of at least 10 μg/mL (~ 30 μM) . This same research group reported on luteolin and wogonin, but their mechanism of action in cervical cancer cells was to suppress HPV18 E6 and HPV16 E6 expression, respectively [28, 29]. Following up on the jaceosidin lead, Kumar, et al. published two virtual screening studies using natural products or a 5000-compound library of compounds with properties similar to jaceosidin [24, 26]. Their docking study identified the E6 residues Phe54, Pro116, Leu117 and Lys122 as jaceosidin interacting residues responsible for the blockade of its interaction with p53.
The Androphy group has reported on designed inhibitors of the HPV16 E6 interaction with the ubiquitin ligase E6 associated protein (E6AP) [19, 31, 32]. Over a dozen flavone derivatives were presented and the 2, 6-substituted benzopyrazone derivative CAF-25 exhibited an IC50 of 2.7 mM in the HPV16 E6/E6AP binding assay and 4.4 mM in a p53 degradation assay with E6. Molecular dynamics coupled with site-directed mutagenesis defined the interactions of CAF-25 with HPV16 E6 residues Lys11, Leu50, Cys51, Leu67, Arg102, and Arg131.
The results for myricetin depicted in Fig. 9 indicate an interaction with the HPV16 E6 residues Leu50, Cys51, Gln107, and Arg131. Binding pocket amino acids that were identified using flavonols in this study do overlap to some extent with the LxxLL binding pocket amino acids [33, 34, 35, 36]. LxxLL peptide inhibitors of E6 are known to induce apoptosis. E6 binding to LxxLL peptides interferes with the apoptotic pathway, leading to cellular proliferation. Small molecule inhibitors binding to the LxxLL binding pocket interfere with the peptide binding, subjecting the cell to apoptosis.
A majority of cervical cancer patients are infected with HPV. Prophylactic HPV vaccines can only reduce or prevent cancers to some extent. These are not effective in already infected patients. Also, HPV vaccines are very expensive and may not be accessible in developing countries. Small molecule inhibitors may be the only choice in unvaccinated countries. In general, HPV16 E6 inhibitors provide a promising option for treating cervical cancers. Though the crystal structure of E6 protein is known, information about specific ligand binding pocket is unknown. In this study, the potential binding pocket that the E6 inhibitors bind was predicted. It was generally observed that Leu50 and Cys51 are necessary for high binding affinity, with Cys51 being essential for hydrogen bonding. Docking Scores of myricetin were higher than the other ligands in most of the binding pockets, which definitely agrees with experimental IC50 values .
Efficacy of the reported ligands or ligands developed for this target may also be improved through conjugation with multivalent glycocalixarenes, which are known to interact with biological macromolecules. Glycocalixarenes have demonstrated reaching viral and bacterial targets and interfere with their infections [37, 38]. E6 drugs could also be combined with other molecules targeting different cancer-specific pathways, such as molecular chaperons like HSP90 and HSP70 to treat cancer and neurodegenerative diseases . For instance, the myricetin and HSP70 activators such as 115-7c were used to investigate the mechanisms underlying in Alzheimer and tau-related diseases .
In summary, identification of the ligand binding pocket helps to design novel inhibitors of the HPV16 E6 oncoprotein as a promising treatment for cervical cancer.
Molecular modeling studies were carried out on a Dell Precision workstation with Intel (R) Xeon (R) CPU E5–1620 v3 @3.50GHz processor. Structure building, docking and analysis were carried out using Discovery Studio (Dassault Systèmes BIOVIA, Discovery Studio Modeling Environment, Release 2017, San Diego: Dassault Systèmes, 2016), GOLD (Genetic Optimization for Ligand Design) suite, version 5.4.0, protein ligand docking package [18, 19, 20], and the PyMOL Molecular Graphics System, Version 1.8 Schrödinger, LLC.
HPV16 E6 crystal structure
The X-ray crystal structure of HPV16 E6 (pdb: 4GIZ) was downloaded from the Protein Data Bank and utilized for the docking experiments [33, 41]. The unit cell contains two instances of the E6 protein designated chain ‘C’ and ‘D’. For the studies described, ‘C’ was used. Preliminary ‘blind docking’, described below, either chain ‘C’ or ‘D’ provided similar results, so the differences between the two instances of the protein were considered to be insignificant.
Blind docking was performed using a web-based program called SwissDock [42, 43] that predicts all potential binding sites on a target protein. SwissDock is also equipped with a database of proteins and ligand structures. It works based on a docking software named EADock DSS . EADock DSS has the ability to dock in a specified region on the protein or all target cavities (blind docking). The protein structure is specified in the program using its PDB code from the Protein Data Bank  or by uploading structure files. The ligand files could be in the Mol2 format or CHARMM formatted files. The ligand can also be selected from the built-in ZINC database  or by uploading the structure. There are three docking settings that can be selected, from very fast, fast, and accurate. The amount of CPU processing time increases along with increasing accuracy. Upon submission of the docking, using the “Submit Docking” tab, it can be followed up using a URL provided. EADock DSS generates several binding modes of a ligand and the most favorable binding modes are clustered and provides an output file called “Predictions” file. Each cluster represents various conformations of a ligand at the given location on the protein and each cluster represents a pocket on the target protein. The different clusters could be viewed in the SwissDock web browser through a Jmol applet. Individual clusters were saved into separate files for further visualization using EditPlus text editing software .
Visualization of protein-ligand interactions
Discovery Studio Visualizer (version 4.1.0) was used to determine protein-ligand interactions. The HPV16 E6 protein and each ligand cluster was opened in Discovery Studio. The amino acid residues that displayed interactions with the ligand were documented. Discovery Studio also shows the types of bonds with different colors and the distance between them.
Docking into individual binding pockets was carried out using GOLD (Genetic Optimization for Ligand Docking) program. GOLD uses genetic algorithm that explores the binding pocket and searches for the best ligand interactions [46, 47, 48]. Docking studies were performed with default settings using 100 genetic algorithm (GA) runs using the GOLD score as a scoring function. Hermes visualizer in the GOLD suite was used to prepare the protein for docking. HPV16 E6 protein was prepared by adding hydrogens, removing extra water molecules and removing metals and other ligands. The receptor binding site was defined by specifying amino acid residues, which were based on the interactions noted from the blind docking in SwissDock. Default values of all other parameters were used. Higher GOLD scores represent greater affinity of the ligand for the protein at the defined binding site.
Conception and Design: SK and JK. Docking, preparation of figures, data analysis, data interpretation, writing, review, and revision of the manuscript: RM, LL, JM, SK, JK. All authors read and approved the final manuscript.
This study is supported by the start-up funds (SK) from Keck Graduate Institute School of Pharmacy for the purchase of Dell precision work station and software GOLD and Pymol. The authors acknowledge the support of The Kenneth T. and Eileen L. Norris Foundation for the Summer Undergraduate Research Experience (SURE) Program (R.M.), and the K12 Internship Program (L.L.) supporting RM and LL respectively who helped running experiments and collecting data.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
- 1.Human Papillomavirus and Related Diseases in the World. Summary Report 27 July 2017 http://www.hpvcentre.net/statistics/reports/XWX.pdf. Accessed 31 July 2019.
- 4.Munoz N, Castellsague X, de Gonzalez AB, Gissmann L. Chapter 1: HPV in the etiology of human cancer. Vaccine. 2006;24(Suppl 3):S3/1–10.Google Scholar
- 5.Munoz N, Bosch FX, de Sanjose S, Herrero R, Castellsague X, Shah KV, Snijders PJ, Meijer CJ, International Agency for Research on Cancer multicenter cervical Cancer study G. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med. 2003;348(6):518–27.PubMedGoogle Scholar
- 28.Ham S, Kim KH, Kwon TH, Bak Y, Lee DH, Song YS, Park SH, Park YS, Kim MS, Kang JW, et al. Luteolin induces intrinsic apoptosis via inhibition of E6/E7 oncogenes and activation of extrinsic and intrinsic signaling pathways in HPV-18-associated cells. Oncol Rep. 2014;31(6):2683–91.PubMedGoogle Scholar
- 34.Ramirez J, Poirson J, Foltz C, Chebaro Y, Schrapp M, Meyer A, Bonetta A, Forster A, Jacob Y, Masson M, et al. Targeting the two oncogenic functional sites of the HPV E6 Oncoprotein with a high-affinity bivalent ligand. Angew Chem Int Ed Eng. 2015;54(27):7958–62.Google Scholar
- 38.Andre S, Sansone F, Kaltner H, Casnati A, Kopitz J, Gabius HJ, Ungaro R. Calix [n]arene-based glycoclusters: bioactivity of thiourea-linked galactose/lactose moieties as inhibitors of binding of medically relevant lectins to a glycoprotein and cell-surface glycoconjugates and selectivity among human adhesion/growth-regulatory galectins. Chembiochem. 2008;9(10):1649–61.PubMedGoogle Scholar
- 45.EditPlus - Text Editor for Windows 4.1 (2016-08-02) https://www.editplus.com/. Accessed 31 July 2019.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.