In-silico designing of a potent analogue against HIV-1 Nef protein and protease by predicting its interaction network with host cell proteins.

BACKGROUND: HIV-1 has numerous proteins encoded within its genome, which acquaints it with the required arsenal to establish a favorable host cell environment suitable for viral replication and pathogenesis. Among these proteins, one protein that is indispensable and ambiguous is the Nef protein. AIM: Interaction of Nef protein with different host-cell protein was predicted and subsequently the down regulation of cluster of differentiation 4 (CD4) was targeted through designing of inhibitors of Nef protein for either preventing or if not at least delaying pathogenesis.
MATERIALS AND METHODS: The interaction network of Nef protein with host-cell proteins were predicted by PIMRider. Analogue of Lopinavir were prepared and evaluated considering all factors affecting the drug stability and toxicity. Finally Docking simulation were performed using an Auto-Dock Tool 4.0.
RESULTS: In the interaction network of Nef protein with different host-cell proteins it was found out that 22 host cell proteins are involved in the interaction and execution of different types of functions in host cell but these functions are altered with the interaction with the Nef protein. After extensive and controlled in silico analysis it has been observed that the analogue LOPI1 binds to Nef protein (2NEF) at CD4 interacting site residues giving minimum binding energy of -7.68 Kcal/mole, low Ki value of 2.34 μM, maximum number of hydrogen bonds (8), good absorption, distribution, metabolism and excretion properties, and less toxicity in comparison with the standard Lopinavir against HIV1 protease (1HPV).
CONCLUSION: The newly designed analogue (LOPI1) is showing significant in silico interaction with Nef protein and protease and can be taken forward as a potent drug lead, which may finally emerge out to be even better than the standard Lopinavir.

HIV-1 is a retrovirus ~9.8 Kb RNA genome.[1] Human Immuno Deficiency Virus type 1 (HIV-1) virus consists two strands of RNA, and each strand has nine genes together code for at least fifteen proteins [Table 1]. On the basis of their primary function, the HIV-1 proteins may be classified into five groups: structural proteins (p24, p17, p7, and p6), envelope proteins (gp 120 and gp 41), enzymes (reverse transcriptase, integrase, and protease), transactivators (Tat, Rev, and Vpr), and regulatory proteins (Vif, Vpu, and Nef).[2-4]

Table 1

Nine Genes of HIV, their proteins and functions

HIV-1 Nef is a small myristoylated protein of 206 amino acids that is indispensable for viral replication and progression of acquired immune deficiency syndrome (AIDS).[56] It contains a structurally flexible N terminal membrane anchor domain of about (1-57) residues and a well conserved folded C-terminal core domain of about (58-203) residues.[7-9] NMR studies have showed that the core of HIV-1 Nef protein contains a type II polyproline helix (70-77aa), followed by two alpha helices (81-120aa), a four stranded antiparallel beta sheets (121-186aa), two additional alpha helices (187-203aa), and the residues (60-71aa) form various flexible solvent-exposed loops [Figure 1].[1011]

Figure 1

(a) Anchor domain of human immuno defiency virus-1 Nef protein (b) Core domain of human immuno defiency virus-1 Nef protein

HIV-1 has no enzymatic activity, therefore its action is based on its interaction with the cellular proteins.[12] The interaction of Nef with CD4 is important because it downregulates the Cluster of differentiation 4 (CD4) and creates a persistent state of infection.[13] The major benefit of Nef-induced CD4 downregulation is probably to promote viral replication with the release of viral particles and it also prevents the super infection event that may lead to premature death of infected host cell.[1415]

CD4 downregulation

Downregulation of CD4 is probably the best explained function of HIV-1 Nef.[16] Nef- induced CD4 downregulation occurs through a two step process: First, it connects the cytoplasmic tail of CD4 with adaptor protein 2 (AP2), thereby stimulating the formation of clathrin coated pits which rapidly internalize the viral receptor.[17] Second, Nef interacts with COPI cotamers in the endosome and targets CD4 for lysosomal degradation.[17-19] NMR spectroscopy has showed that the 13 residue peptide (MSQIKRLLSEKTT) from the cytoplasmic tail of CD4 binds to HIV-1Nef protein and this part is crucial for down regulation of CD4 by Nef.[10] The residues of HIV-1 Nef which are involved in associating with CD4 are W57, L58, G95, G96, L97, R106, and L110 and are present in the Nef core domain [Figure 2].[1011]

Figure 2

(a) Shows peptide residues of Cluster of differentiation 4 cytoplasmic tail binds with Nef (b) Shows amino acid residues of Nef interacts with Cluster of differentiation 4

Lopinavir

Lopinavir is an anti-retroviral drug under the umbrella of protease inhibitors, which is used for hindering the activity of the virus. In the work presented herein lopinavir has been taken forward for the designing of its analogue [Figure 3a]. It is marketed by Abbott as Kaletra, a co-formulation with a sub-therapeutic dose of ritonavir, as a component of combination therapy to treat HIV/AIDS. It is available in tablets as well as in oral solutions. Tablets can be taken without meals and oral solution can be taken with food. Oral tablet is stable at room temperature and oral solution is stable at 2°C to 8°C until date on label and is stable when stored at room temperature (up to 25°C or77°F) for 2 months.[20]

Figure 3

(a) Structure of Lopinavir (b) Structure of human immuno defiency virus Protease with active site

HIV protease

HIV protease is an essential enzyme primarily for the proper assembly and maturation of virus. It acts as a scissor, which cuts in the inactive polypeptide string into the active proteins for each virion. It contains two identical chains of 99 amino acids with characteristic active site sequence of D-T-G at position 25-26-27, which is C2-symmetric in its free form [Figure 3b]. Each monomer has an extended β-sheet (GLY rich loop) known as flap domain that constitute in part the substrate binding and one of the two essential aspartyl residue, Asp25, which appears on the bottom of the cavity.[21] Active site pocket of protease (1HPV) remains highly hydrophobic (hydrophobic score 1.732), which facilitates the process of drug binding.[22]

Materials and Methods

Proteins selected for this study

For this study of HIV-1 Nef protein, two separate protein domains were retrieved; first domain is 1QA4 (Nef anchor domain) and second domain is 2NEF (Nef core domain). For CD4, 2KLU (cytoplasmic tail of CD4), and for protease, 1HPV (unmutated) have been retrieved from RCSB Protein Data Bank (http://www.rcsb.org). Also, 1HPV and 2NEF are used for further analysis.

Structural study of proteins

For the structural study of proteins, the PyMOL software (Developed by- Warren Lyford DeLano and commercialized by DeLano Scientific LLC, South San Francisco, California, U.S.A) has been used. It helps in the visualization of the protein PDB's, removal of non- essential molecules (like water and ligand) from protein PDB files and analysis of docking results. The position of alpha helix, beta sheets, interacting peptides or residues and active site pocket has been studied upon through PyMOL.

Prediction of interaction network

PIMRider, the hybrigenics functional proteomic software (http://pim.hybrigenics.com/) is a web application dedicated to exploration of protein pathway and protein interaction maps (PIM) through sophisticated visualization tools. PIMRider is used herein for predicting the interaction network between HIV-1Nef and human host cell proteins (By-Register-Entry points-HIV-human interaction).

Analogue preparation

In the work presented herein Lopinavir, a HIV-1 protease inhibitor, is used as a parent scaffold for the efficient and effective designing of potent analogue. Mainly three steps have been considered in designing the analogue (LOPI1) of Lopinavir drug.

In the complex structure of Lopinavir and 1HPV, those amino acid residues were visualized which make the interaction and are present around the Lopinavir (D25, T26, G27, A28, D29, D30, T31, V32, M46, I47, G48, G49, I50, T80, P81, V82, N83, I84) in the active site pocket.

In HIV-1 Nef (2NEF) protein, those amino acid residues were visualized which play an important role in binding with CD4 (W57, L58, G95, G96, L97, R106, L110) and form a pocket-like structure.

Atoms and groups of Lopinavir have been modified on the basis of their size, orientation and properties of amino acids residues, which are present in both proteins.

DrugBank database (http://www.drugbank.org) is a unique bioinformatics and chemo informatics resource that contain all the information of drugs. The PDB and MOL file of lopinavir was retrieved from drug bank database.[23]

ACD/ChemSketch is the powerful all-purpose chemical drawing and graphics package from ACD/Labs developed to help chemists to quickly and easily draw molecules, reactions, and schematic diagrams and calculate chemical properties. ChemSketch is adopted in this study for the designing of analogue (LOPI1) of Lopinavir, generating its SMILE notation and retrieving the chemical formula of the designed analogue. Smile format of LOPI1 (Analogue) was uploaded in an online tool CORINA (http://www.molecular-networks.com/online_demos/corina_demo) for generating the 3D structure. CORINA is a widely used online program for the conversion of 2D format into 3D format of ligand.

In silico absorption, distribution, metabolic and excretion and toxicity prediction

Smile formats of Lopinavir and LOPI1 (analogue) were submitted in an online tool absorption, distribution, metabolism and excretion (ADME)/Tox WEB (http://pharma-algorithms.com/webboxes/). This tool predicts out ADME and Toxicity of both the drug and its derivative. Another online source ChemSilico Secure Web Server (https://secure.chemsilico.com/index.php.) was used for the prediction of Blood Brain Barrier (BBB) and AMES Test Mutagenic Index (AMI). Lipinski rule of five was performed with the help of Lipinski filter facility available online at Supercomputing Facility for Bioinformatics and Computational Biology, Indian Institute of Technology, New Delhi, and India.[24]

Docking simulation

Docking experiments were performed using an Auto-Dock Tool 4.0, suite of automated docking tools. It is a protein ligand docking tool in which a genetic algorithm is used to find the binding conformation of a ligand. It is most commonly used docking tool adopted and referred to in the scientific literature. The Docking process involves four main steps: (1) Receptor preparation, (2) Ligand preparation, (3) Docking using a search algorithm, (4) Analysis of binding conformation of ligand. In the work presented herein docking has been done between Lopinavir and 1HPV (Unmutated), LOPI1 (Analogue), and 1HPV, LOPI1 (Analogue) with 2NEF. First of all, the non essential molecules like water and ligand were removed from both protein PDB files, the energy was minimized of both the protein and ligand followed up by saving them in a pdbqt format. The Grid has been set at the centre of active site pocket, which covers all the residues present inside the active site pocket with 52*46*66 points in x, y, z direction and –2.028, –3.722, 0.194 grid centre for 1HPV[22] and another grid has been set at with 64*64*64 points in x, y, z direction and 1.972, 8.278, 0.417 grid centre for 2NEF. The parameters were saved as grid parameter file (.gpf) and followed by Autogrid run. In third step of docking, parameter files (.dpf) were prepared in which genetic algorithm has been selected, the value of which remains as default. These values determine the optimal run parameter which depends upon the nature of ligand and proteins (receptor). Ten generations were set for each GA run. This was followed by saving the parameters as docking parameter file (.dpf) and finally subjected to Autodock run. Docking simulation has been repeated for three times with similar parameters to improve the precision level of results. Care should be taken that similar results should be obtained after each repeats. The results generated were visualized in PMV (Python Molecular Viewer) and PyMOL. The interactions were studied in terms of minimum binding energy (Kcal/mol), Ki (Inhibition constant) value in μM, and number of hydrogen bonds formed between the active site residues of macromolecule and ligand.

Results

Interaction network between HIV-1Nef and human host cell proteins

Protein interaction map is the most comprehensive interaction map based on the relation between the HIV and its host cell in humans that is the lymphocytes. The predicted interaction network showed the interaction of HIV-1 Nef protein with CD4 and also with the other host cell proteins [Figure 4].

Figure 4

Interaction network of human immuno defiency virus-1Nef protein with other host cell proteins

The Nef protein interacts with many other human host cell proteins for working properly in the host cell. There are 22 host cell proteins, which participate in the interaction network of Nef protein. The name and function of all 22 host cell proteins has been given below [Table 2].

Table 2

Name and functions of interacting host cell proteins

ADME and toxicity test of lopinavir and derivative (LOPI1)

Absorption distribution metabolic excretion (ADME) properties and toxicity of Lopinavir and LOPI1 (Analogue) have been predicted with the help of reliable online tools as discussed above. In this study, it has been observed that the newly designed derivative LOPI1 showed good ADME properties and very low toxicity pattern than the previously available standard drug Lopinavir [Table 3]. LOPI1 does not show any mutagenic effect rather than it showed good protein binding property.

Table 3

Predicted ADME properties and toxicity values of lopinavir and LOPI1

Oral bioavailability and toxicity properties were predicted by ADME/Tox WEB (http://pharma-algorithms.com/webboxes/). BBB, Ames test Mutagenic Index (AMI) was predicted by ChemSilico Secure Web Server (https://secure.chemsilico.com/index.php).

Docking analysis

Docking simulation of Lopinavir and LOPI1 (Analogue) with HIV Protease (1HPV) and HIV-1 Nef (2NEF) was performed and the results were expressed in terms of minimum binding energy (Kcal/mole), Ki value (inhibitory constant), and number of hydrogen bonds formed. After the docking analysis it has been observed that the LOPI1 (Analogue) shows –4.78 Kcal/mole minimum binding energy, 315.03 μM Ki value, and 5 hydrogen bonds [Figure 5a and b] with 2NEF. Further, Lopinavir showed –5.73 Kcal/mole minimum binding energy, 63.29 μM Ki value and 5 hydrogen bonds [Figure 5c and d] with 1HPV. But when the docking results between LOPI1 (Analogue) and 1HPV was carried out, it was found out that LOPI1 showed better interaction results than standard Lopinavir. LOPI1 shows least minimum binding energy of –7.68 Kcal/mole and 2.34 μM Ki value with 8 hydrogen bonds [Figure 5e and f], which were found out during second run out of the ten docking runs. On the basis of binding with 2NEF (at desired residues) and 1HPV (HIV Protease), we can probably suggest keeping in mind the finding of least minimum binding energy, less Ki value and number of hydrogen bonds [Table 4] that LOPI1 can be taken forward as a more effective and efficient drug lead against HIV Nef protein and protease and is proposed to be even better than the standard Lopinavir.

Figure 5

(a) Docked structure of LOPI1 and 2NEF, visualized in PyMOL (Yellow dotted line show hydrogen bonds) (b) Docked structure of LOPI1and 2NEF, visualized in PMV (c) Docked structure of Lopinavir and 1HPV, visualized in PyMOL (d) Docked structure of Lopinavir and 1HPV, visualized in PMV (e) Docked structure of LOPI1 and 1HPV, visualized in PyMOL (f) Docked structure of LOPI1 and 1HPV, visualized in PMV

Table 4

Docking studies of lopinavir and LOPI1 with HIV Protease (1HPV) and HIV-1 Nef (2NEF)

Discussion

The brutal attack on humanity by HIV-1 has been already proven to be afflictingly difficult to hinder. For blunting this epidemic, antiretroviral's (ARVs) have been developed that inhibit and play a crucial role in HIV-1 infection. But unfortunately the HIV-1 has an ability to mutate and it requires multiple drug treatments, which are limited in their application by their side effects and expenses.[2526] So there is a need to consider all possible targets to counter HIV-1. Because of having a central role in HIV pathogenesis the HIV-1 Nef protein has been proposed and considered for this work. It has been observed that the long term survivors of HIV infection who show lack of disease progression are usually associated with Nef deleted gene, which defines the necessity of Nef to act as a pathogenic factor.[27] There are varieties of functions carried out by Nef, which are based on its ability to interact with multiple cellular proteins. On the basis of interaction network between HIV-1 Nef and the host cell proteins, it has been observed that there are 22 proteins which interacts with the viral Nef protein and perform different types of functions, that are essential for the life cycle of the virus [Figure 4]. Thus, we can say that the Nef protein acts as a adaptor protein, that divert the host cell protein from the right path and performs those functions which amplify viral replication and pathogenesis. The name and functions of all 22 host cell proteins in a cell have been described [Table 2]. Nef performs mainly four functions:-

CD4 downregulation

MHC-I downregulation

Activation of p21-activated protein kinase (PaK2)

Enhancement of viral particle infectivity.[28]

Each of these four Nef functions could serve as contributors to Nef's deceptive role in replication and disease progression.[27] Several reports have suggested the importance of removal of CD4 from the infected cell surface for production of infectious virus particles.[2930] Without this function of Nef, the CD4 can bind to Env protein during virion budding and interfere with the production of fully infectious viral particles.[27] The downregulation of CD4 by Nef protein enhances the virus replication by preventing the super infection events, that would lead to premature death of infected host cell and this function of Nef also represents a strategy to control signalling events in the infected cell.[6] CD4 can enhance antigen-driven TCR signalling incidents through interaction between its extracellular domain and MHC class II molecule, with the union of Lck tyrosine kinase with CD4 cytoplasmic domain. The effect of Nef on CD4 is that it may interfere with TCR signalling to the benefit of virus, possibly by decreasing the activation of induced apoptosis.[31-33]

Lopinavir is a well-known protease inhibitor, which is used to block the HIV protease enzyme thereby inhibiting its activity. The pocket area of the active site plays an important role in the drug packaging. In wild type protease, active site pocket area is 581.9 Å with 965.9 volume (ID 21). Flap domains are presents at floor of the active site pocket and it is observed that the closed conformation of the flap helps in maintaining the geometry of active site binding pocket area as well as it provides good support to the drug at the time of binding. Active site pocket provide good hydrophobic environment (H.P. score = 1.732) for better interaction of drugs. It is also well known that active site also plays an important role in the drug packaging.[22]

At the time of designing of potent analogue LOPI1,[22] some modifications were made on the basis of position and properties of surrounding residues in 1HPV as well as active site residues in 2NEF by using an offline Chem Sketch software [Figure 6]. In order to determine the conformation of the receptor-ligand complex, docking has been done by choosing a ligand conformation with best binding affinity value (Kcal/mol) or the smallest one.[34-36] After this analysis it was observed that our modifications in the standard Lopinavir have resulted in good interaction with active site residues in 2NEF as well as active site pocket residues in 1HPV. The modification LOPI1/O37 interacts with GLY95/O in 2NEF and LOPI1/O1 with GLY48/Ob, LOPI1/O37 with PRO81/Oa and LOPI1/O41 interact with GLY48/Na in 1HPV through hydrogen bonds. LOPI1 also showed good ADME properties and less toxicity than standard Lopinavir. The docking results of LOPI1 with Nef protein (2NEF) gave 5 hydrogen bonds with GLY95/O-LOPI1/O37, LEU58/O-LOPI1/O28, ARG106/NH1-LOPI1/O9, ILE59/O-LOPI1/O6, and ALA60/O-LOPI1/O22 as well as it also showed interaction with GLY96/O-LOPI1O9, GLY96/O-LOPI1O6 in other run (3). The docked result of Lopinavir with HIV Protease (1HPV) was observed and it was found out that 5 hydrogen bonds were formed with ASP29/Ob-LOPI1/O47, ASP29/Nb-LOPI1/O47, GLY27/Ob-LOPI1/O93 ASP25/Ob-LOPI1/O93, and GLY27/Oa-LOPI1/O71. When the docked results of LOPI1 and 1HPV were visualized it showed 8 hydrogen bond with GLY48/Ob-LOPI1/O1, ASP25/Oa- LOPI 1/O6, GLY27/Oa- LOPI 1/O9, GLY27/Oa- LOPI 1/O22, ASP29/Na- LOPI 1/O22, ARG8/NH2b-LOPI 1/O28, GLY48/Na- LOPI1/O41, and PRO81/Oa-LOPI1/O37 [Table 4].

Figure 6

2D structure of LOPI1 (Analogue)

LOPI1 clearly showed that it binds with the CD4 interacting site residues in Nef (Gly95, Gly96, Leu58, Arg106) as well as it also gives least minimum binding energy (–7.68 Kcal/mole), least Ki value (2.34 μM) and maximum number of hydrogen bonds (8) with receptor sites of protease (1HPV). Least minimum binding energy of LOPI1 than standard lopinavir clearly indicates that the rate of association of drug molecule with the receptor site is increased,[3738] least Ki value (concentration of drug required for inhibition) indicates that the lower concentration of LOPI1 is required for inhibition than lopinavir, maximum number of hydrogen bonds with receptor site (1HPV) further suggested and highlighted the strong binding of LOPI1 with receptor site than Lopinavir [Table 3].[39] LOPI1 also shows the best binding pattern inside the active site pocket of 1HPV than lopinavir [Figure 5e–f]. Determination of toxicological properties for molecules in the development of drug designing is getting more important. One method that is used in this toxicological study is ADME/Tox.[39] Similarly this method for developing an insilico new potent drug has also been adopted in the recent years by other researchers involving the designing of an analogue of an existing drug.[40-42] On the basis of ADME and docking simulations, it is strongly suggested and proposed that our designed and in silico tested analogue LOPI1 will probably work in two different directions, as a good protease inhibitor as well as interrupting the function of HIV-1 Nef CD4 downregulation.

Conclusion

On the basis of extensive and comprehensive studies of the structure and function of HIV-1 Nef protein it may be concluded that even though a small protein it can do a lot of damage to the immune system of an infected cell. The predicted interaction network of Nef clearly indicates that it interacts lonely with dozens of cellular components and performs its functions which is essential and vital for viral life cycle. On the basis of results obtained through insilico analysis it is observed that the designed novel and potent analogue of lopinavir (LOPI1) works with two beneficial traits; firstly it shows better binding with CD4 interaction site in Nef and secondly it also increases the moral goodness of Lopinavir with HIV protease by showing efficient and effective results like least minimum binding energy, least Ki value, maximum number of hydrogen bonds, and best binding pattern. It was also found out that LOPI1 showed good ADME properties and less toxicity than standard Lopinavir. Finally, on the basis of insilico analysis we can conclude that our study may be serviceable for the development of new antiretroviral drugs, which will be potentially useful in prophylaxis for AIDS epidemic. Additionally, this study also opens up the hidden potential of cheap, fast, and reliable bioinformatics tools and softwares for the exploration of mechanism of action of standard drugs and their analogues before moving on to the costly and time-consuming wet lab studies.