- Email: [email protected]
Phylogenetic analysis, structure modeling and docking study of HCV NS3 protease for the identification of potent inhibitors
Infection, Genetics and Evolution 59 (2018) 51–62
Contents lists available at ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Research paper
Phylogenetic analysis, structure modeling and docking study of HCV NS3 protease for the identification of potent inhibitors Asad Ziaa, Sumra Wajid Abbasib, Shabeer Ahmadd, Muhammad Ziaa, Abida Razac,
T
⁎
a
Department of Biotechnology, Quaid-i-Azam University, Islamabad, Pakistan Department of Bioinformatics, Quaid-i-Azam University, Islamabad, Pakistan c National Institute of Lasers and Optronics, Islamabad, Pakistan d Department of Microbiology and Biotechnology, Abasyn University Peshawar, Pakistan b
A R T I C L E I N F O
A B S T R A C T
Keywords: NS3 helicase Sequencing NS5B polymerase inhibitors Homology modeling Molecular docking
The nonstructural protein 3 (NS3) helicase of HCV is believed to be a plausible target for the identification and designing of potent antiviral drugs. NS3 protein is involved in a positive sense single-stranded viral replication as well as it also cleaves viral poly protein into diverse mature proteins at different time spans. Structural exploration of NS3 revealed that HCV helicase could also act as translocase. In order to identify potential inhibitors for HCV-3a, the current study has been designed. Serum samples from the Pakistani HCV positive patients were collected, sequenced and after purification included in the present study. Phylogenetic analysis on the samples clustered around it in the same group with those from India. Using homology modeling technique, we determined 3D structure of NS3 gene of HCV-3a and employed further in docking studies to discover potent inhibitor against it. As a result of docking Compound 1, with IC50 value of 0.015 and −14.4 kcal/mol energy, ranked as a most pungent inhibitor among all the studied inhibitors. Compound 1 also exhibited good hydrogen bond interactions with the modeled protein. The finding of present study could be used as a lead in future to design an effective dual inhibitor against HCV-3a.
1. Introduction Hepatitis C virus (HCV) belongs to Flaviviridae, is a blood-borne pathogen which infects about 180 million individuals throughout the world; most of the cases lead to end stage liver diseases, fibrosis, cirrhosis and hepatocellular carcinoma (Choo et al., 1989; Major and Feinstone, 1997; Wasley and Alter, 2000). About 10 million populations of Pakistan are infected with HCV, with rare cases of spontaneous clearness; most of them progress to chronic cases (Ali et al., 2016). The most prevalent genotype among eleven HCV genotypes in Pakistan is 3a (Waheed et al., 2009). For a decade pegylated interferon a (PEG-IFN-a) plus ribavirin (RBV) remained treatment of the choice, is expensive, associated with severe side effects and effective for certain genotypes (Dillon, 2004; Farci et al., 2006; Ferenci, 2006; Ni and Wagman, 2004; Reichard et al., 1998). Different treatment response pattern to HCV genotypes is due to the genetic heterogeneity of virus. Considerable variability is shown by the virus which directly interferes with the disease treatment. The response to treatment varies according to HCV genotype and subtype (Bastos et al., 2016). The sustained virological response rate (SVR) in Pakistani population infected with 3a genotype to IFN-α and RBV combination therapy is 87.5%, approximately 2.45% ⁎
of the patients discontinued treatment due to adverse side effects (Ali et al., 2016). Lacking of efficient treatment regimens and increased incidence rate of HCV infection has created a pressure for the therapeutic compounds that can efficiently target the HCV (López-Labrador, 2008). Nonstructural protein NS3/NS4A serine protease and helicase are considered as potential drug targets for the development of effective anti-HCV compounds (Ashfaq et al., 2011). The main role of NS3/NS4A is to cleave viral poly protein into different mature proteins at various time intervals as well as involved in viral replication; HCV helicase affects the viral life cycle at two steps for unwinding of double strand RNA intermediate required for the movement of HCV NS5B polymerase (Piccininni et al., 2002). NS3 structural analysis revealed a new function of HCV helicase as translocase and considered as a potential specific inhibitor to block NS3 helicase (Gu and Rice, 2010). For predication and comparison of molecular and physicochemical properties and mechanisms of reactions of different therapeutic compounds, molecular modeling techniques are widely used (Elfiky et al., 2013; Saleh et al., 2014). For HIV-1 and HCV proteases effective inhibitors are designed through these techniques (Elfiky et al., 2013; Ibrahim et al., 2012a; Ibrahim et al., 2012b). Computational studies not only motivated researchers for the identification of novel therapeutic targets but also
Corresponding author at: Nanotheranostics Research Group, National Institute of Lasers & Optronics (NILOP), Islamabad, Pakistan. E-mail address: [email protected] (A. Raza).
https://doi.org/10.1016/j.meegid.2018.01.026 Received 9 November 2017; Received in revised form 24 January 2018; Accepted 28 January 2018 Available online 31 January 2018 1567-1348/ © 2018 Elsevier B.V. All rights reserved.
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
fortify the drug designing/development process. The field of computational biology and its latest uses has offered reliable methods for the enhanced understanding of biological systems of interest (Azam et al., 2009a; Azam et al., 2009b, 2009c; Azam et al., 2010; Azam et al., 2012; Barreca et al., 1999; Cheng et al., 2002). In the current study, the potent NS3-3a protease inhibitors, previously designed for NS3, have been identified by employing different computational techniques such as homology modeling, and molecular docking. 2. Methods 2.1. Complementary DNA synthesis and amplification of NS3 gene HCV positive patients with genotype 3a from the underprivileged part of Islamabad were collected and included in the present study. RNA was extracted from the serum samples using Virus RNA isolation Kit (INSTANT, AJ Roboscreen, GmbH Germany). The RNA was reverse transcribed in to complimentary DNA (cDNA) using 100 units of M-MLV reversed transcriptase (Invitrogen), with 5 pM of outer antisense primer 5-GGCGACACTCCACCATAGAT-3. Amplification of NS3 gene using sense primer 5-GGCCGTGAGGTGTTGTTGG-3 and anti-sense primer 5TGGTTACTTCCAGATCGGCTG-3 was carried out according to Sabri and colleagues (Sabri et al., 2014). After amplification and purification of the PCR product sequencing was carried out by Sanger method. 2.2. Phylogenetic analysis The sequences obtained from Sanger method were subjected to a phylogenetic analysis for determination of regional distribution of NS3. HCV NS3 sequences from the regions of UK, Germany, France, Australia, India, Nepal, Canada and Somalia were extracted from NCBI database. Alignment and phylogenetic tree construction has been carried out using MEGA 7.0. For multiple sequence alignment clustalw algorithm was used (Sievers et al., 2011). The tree was constructed using neighbor joining algorithm (Fig. 1). Neighbor-Joining method was used in order to infer the evolutionary history (Saitou and Nei, 1987). The optimal tree along with branch length sum equal to 1.88376007 is presented. The phylogenetic tree was drawn to scale in the same branch length units as of evolutionary distance. Maximum Composite (Tamura et al., 2004) Likelihood method was employed to compute evolutionary distances in the number of base substitution per site units. Total sequences involved in the analysis amounted to 32. Missing data and all the positions presenting gaps we eliminated. Final dataset comprised of a total of 1850 positions. Further, MEGA7 was used to conduct evolutionary analysis (Kumar et al., 2016). Fig. 1. 2.3. Homology modeling The crystallographic three-dimensional structure of HCV NS3 genotype 3a, the genotype most prevalent in Pakistani community, has not been reported yet. The sample used in the present study was collected from the underprivileged part of Islamabad, after sequencing the sample was designated KP8/NS3 sequence, in FASTA format. In order to identify suitable templates from RCSB PDB, NCBI BLASTp search of a target sequence was accomplished (http://www.pdb.org/pdb/) (Altschul et al., 1997). BLAST searching resulted in the most appropriate template (PBDID 4B6E) with a resolution of 2.64 Å from Hepatitis C virus (Saalau-Bethell et al., 2012). To generate a 3D model by alignment by means of Clustal Omega, coordinates of 4B6E were used as a template (http://www.ebi.ac.uk/Tools/clustalw2/index.html) (Sievers et al., 2011). To characterize the areas of similarity, pair-wise sequence alignment was performed. This characterization might emphasize evolutionary, functional, structural relationships among target and template two. In total, 5 homology models of NS3 for genotype 3a, with differing geometric conformations, were generated by using MODELLER 9v12
Fig. 1. Phylogenetic analysis of the HCV isolates from the underprivileged part of Islamabad representing the regional distribution of these isolates.
52
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 2. Alignment of HCV NS3 sequence and template 4BE6.Sterics are indicating the conserved residues.
Table 1 PROCHECK, ERRAT, VERIFY_3D and Z-scores for models generated by MODELL. Models
PROCHECK
MODELLER
Residues in core Region (%)
Additionally allowed Region (%)
Generously allowed Region (%)
Disallowed Region (%)
Bad contacts
Model1
481 92.1% 485 92.9% 484 92.7% 479 91.8% 486 93.1%
37 7.1% 33 6.3% 34 6.5% 37 7.1% 31 5.9%
2 0.4% 3 0.6% 3 0.6% 4 0.8% 3 0.6%
2 0.4% 1 0.2% 1 0.2% 2 0.4% 4 0.4%
−0.4 inside −0.3 inside −0.7 inside −0.3 inside −0.4 inside
Model2 Model3 Model4 Model5
ERRAT
Verify_ 3D
ProSA-web
G factor
Quality factor
3D-1D score > 0.2 (%)
Z-score
1.2 better 1.1 better 1.2 better 1.1 better 1.2 better
81.06
93.32%
−8.6
81.09
93.00%
−8.7
77.43
92.21%
−8.5
81.52
91.10%
−8.6
80.51
91.73%
−8.7
score affirms the reliability and validity of studied model. Keeping in mind the end goal to check similarity of modeled protein by arrangement of its own amino acids, Verify3D was utilized. A direct correlation between the value of Verify3D and quality of model exists, higher the value, more better will be the quality. Root mean square deviation (RMSD) values were calculated for the verification of the symmetry of template and target proteins.
(Šali et al., 1995). Among the generated models, the best model was chosen on the basis of PROCHECK analysis (Laskowski et al., 1993). The selected model was then subjected to energy minimization using UCSF Chimera (Pettersen et al., 2004). Gasteiger–Huckel charges were also assigned to protein. A structural evaluation of minimized structure was carried out by using ProSA-web (Wiederstein and Sippl, 2007), Verify3D (Eisenberg et al., 1997) and ERRATv2.0 (Colovos and Yeates, 1993). The stereochemistry of crude model was checked using Ramachandran plot, which is part of PROCHECK program. Energy profile of selected model was obtained by ProSA test as Z-score. Negative Z-
53
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 5. Z-score analysis of Model 5 using ProSa web. Fig. 3. Homology model of HCV NS3 (Pakistani strain 3a) generated using MODELLER.
approved protease inhibitors such as telaprevir and boceprevir were also used (Lin et al., 2006; Malcolm et al., 2006; Perni et al., 2006). Chem Office 2004 was used to draw the bi-dimensional structure of inhibitors. The two dimensional structures of inhibitors were drawn using Chem Office 2004 (Li et al., 2004). Chimera was employed to reduce the energies of selected inhibitors and further used for docking.
2.4. Active site prediction CastP online server was utilized for the detection of active site in the predicted protein model (Dundas et al., 2006). In order to identify the conserved residues the sequence of modeled protein was aligned with the template's sequence (4B6E) using Clustal omega Fig. 2.
2.6. Molecular docking protocol 2.5. HCV inhibitors Graphical User Interface program Auto Dock Tool (ADT) was utilized for the preparatory steps, for instance preparation of pdbqt files for ligands and protein as well as grid box creation (Morris et al., 2009). Auto Dock assigned united atom Kollman charges, polar hydrogen and fragmental volumes to the protein with solvation parameters. The
Keeping in view the possible application of NS5B polymerase inhibitors in combination with NS3 proteases inhibitors to treat HCV, we utilized di-inhibitors reported active against NS5B 3a by Azam and coworkers in the current study (Azam et al., 2014). Additionally, two FDA
Fig. 4. ERRAT plot of best Model 5.
54
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 6. Superimposed 3D structure of target (NS3) and template (4B6E). Target is shown in dim grey whereas template is shown in purple. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 7. CastP predicted active site for HCV NS3. The active site region has been highlighted in green. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
prepared file was saved by ADT in pdbqt format. The grid map was prepared using a grid box utilizing Auto Grid. The grid size was set to 58 × 66 × 48 xyz points with grid spacing of 0.375 Å while at dimensions (x, y, and z): 4.752, 11.341 and −36.749, grid center was placed. The computational time was limited by ascertaining the scoring grid from the ligand structure. Utilizing the protein and ligand information, Auto Dock-Vina was used for docking along with grid box properties. During docking, both protein and ligands were assumed as rigid. The results < 1.0 Å were clustered together in positional root-mean square deviation (RMSD), and represented with the most favorable free energy binding. The pose with lowest energy of binding or affinity was extracted and further aligned for further analysis with the studied protein.
across the world and 4 sequences were used from samples of HCV extracted from patient of underprivileged part of Islamabad, Pakistan. Clustering of these sequences represent the geographical barriers between these regions. Our sample sequence were found to be clustering with the Indo-Pak peninsula while the rest of sequences where grouped according to continents. Only one of the UK sequence showed abnormality in clustering which may have arisen by constant travelling of individual from Indo-Pak peninsula to Great Britain. From our analysis we can confirm that these are specific to Indo-Pak peninsula and further drug therapy on these sequences would be more beneficial than a generalized therapy around the world.
3.2. Homology modeling and model validation 3. Results and discussion The availability of 3D structure is very crucial in order to explore the potent inhibitors and to study their interactions (Sali and Blundell, 1993). The structure of NS3 of HCV genotype 3a has not been predicted yet, so it created a need for the development of its homology model.
3.1. Phylogenetic analysis Analysis was carried out using 28 different regional sequences from 55
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Table 2 Binding affinities and IC50 values of Top 10 active compounds and FDA approved NS3 protease inhibitors against NS3. S. No.
Compounds
1
IC50 (μM)
Binding affinities
0.015
−14.4
0.014
−13.5
0.075
−13.4
0.024
−13.3
0.019
−13.1
0.038
−12.7
17 of 3-Heterocyclylquinolone
2 16 of 3-Heterocyclylquinolone
3 14 of 3-Heterocyclylquinolone
4 15 of 3-Heterocyclylquinolone
5 13 of 3-Heterocyclylquinolone
6 18 of 3-Heterocyclylquinolone
(continued on next page)
56
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Table 2 (continued) S. No.
Compounds
IC50 (μM)
Binding affinities
1.6
−12.5
0.023
−12.3
Telaprevir
< 0.2
−11.7
Boceprevir
0.5
−10.4
7 7 of 3-Heterocyclylquinolone
8 11 of 3-Heterocyclylquinolone
indicating the good quality and suitability of the generated model. In the next step, to further optimize the Model5energy minimization of was accomplished. The minimized structure of the best model is provided in Fig. 3. In order to evaluate the compatibility of generated model and amino acid sequence, the best model was assessed by means of various verification web tools such as ERRAT, ProSA and Verify3D. The overall quality factor estimated via ERRAT was 80.51% for KP8, 3a NS3, showing the good quality of structure (Fig. 4, Table 1). The Z-score of −8.7, for the best model, was obtained using ProSA web server (Fig. 5, Table 1). Lastly, Verify3D was used to calculate the packing quality of every amino acid. By making use of a scoring function, it made a comparison between residues and their environment in generated models and assigned a 3D–1D score (> 0.2) for each and every modeled protein. In current case 91.73% of the residues of Model 5 occupied an averaged 3D-1D score greater than and equal to 2 (Table 1). Thus, affirming the
The template selection is thought to be the most important step in homology modeling as the accuracy of model depends on it. NS3/4a protein from Hepatitis C virus (PDB entry: 4B6E) was selected as a template, with 100% of sequence coverage, 78% identity and e-value 0.0. Higher similarity between template and target might be a plausible reason for the similar geometric arrangement. So sequence alignment confirmed the protein conservation on sequence level. Homology models for query sequence were predicted using protein modeling tool MODELLER. MODELLER generated 5 models among which the best model was selected on the basis of Ramachandran Plot, generated by PROCHECK. A comparison of models generated by MODELLER, based on overall stereochemistry, illuminated that model 5was best among all. Model 5exhibited good percent of allowed, disallowed and general region in comparison with other models in Table 1. Ramachandran plot of the Model5 showed that 93.1% of the residues lied in the most favored regions, whereas 0.4% of the residues lied in the disallowed regions. Overall G factor value, 1.2, was also in acceptable range 57
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 8. 3D representation of compund1-NS3 docked complex generated using Chimera. The inset represents magnified view of binding pocket. Fig. 9. 2D depiction of compund1-NS3 docked complex using LIGPLOT. Dashed line = hydrogen bond interaction.
58
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 10. 3D representation of Telaprevir NS3 docked complex generated using Chimera. The inset represent magnified view of binding pocket.
Fig. 11. 3D representation of Boceprevir NS3 docked complex generated using Chimera. The inset represent magnified view of binding pocket.
Gly29, Glu30, Val31, Gln32, Val33, Leu34, Ser35, Thr36, Ala37, Thr38, Gln39, Thr40, Phe41, Leu42, Thr45, Met50, Tyr54, His55, Gly56, Ala57, Gly58, Arg60, Thr61, Leu62, Ala63, Gly64, Val65, Lys66, Pro68, Ala69, Val76, Asp79, Trp83, Pro84, Ala85, Pro86, Pro87, Gly88, Ala89, Leu92, Val105, Thr106, Arg107, Ala109, Lys134, Pro140, Met142, Arg153, Pro203, Thr204, Gly205, Gly207, Lys208, Ser209, Thr210, Asn227, Pro228, Ser209, Val230, Ala231, Ala232, Gly235, Phe236, Ser238, Phe239, Arg242, Gly253, Asn254, Thr267, Tyr268, Gly269, Lys270, Asp288, Glu289, His291, Ala292, Gln293, Asp294, Ser295, Thr296, Ser297, Ala321, Thr322, Pro323, Pro324, His331, Asn333, Leu334, Ile367, Arg369, Lys370, Met373, Tyr389, Tyr390, Arg391, Gly392, Leu393, Asp394, Val395, Ile398, Thr400, Thr409, Asp410, Ala411, Met413, Thr414, Gly415, Phe416, Thr417, Gly418, Ala429, Val430, Glu431, Gln432, Tyr433, Val434, Phe436, Glu445, Thr446, Arg447, Thr448, Ala449, Pro450, Asp452, Val454, Ser455, Ser457, Gln458, Arg459, Arg460, Gly461, Arg462, Thr463, Gly464, Arg465,
accuracy and compatibility of selected model with its sequence. Further, the topology of the target and the template protein was verified by superimposing both the structures using Chimera which gives the root mean square deviation (RMSD) value of 0.59 Å (6).When the structure of minimized protein was closely observed, it depicted the same structural topology as of the template. Template protein (4B6E) consists of two domains: protease domain and helicase domain. The concurrence of the said domains in a protein has a plays a vital role maintaining the affinity, selectivity and catalytic rate for their particular substrates (Saalau-Bethell et al., 2012) (Fig. 6). 3.3. Active site prediction The important active site residues identified by Cast Pare Ile1, Thr2, Ala3, Tyr4, Ala5, Gln6, Gln7, Thr8, Arg9, Arg10, Leu11, Thr14, Ile15, Thr17, Ser18, Leu19, Gly21, Asp23, Asn 25, Val26, Val27, Ala28, 59
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 12. 2D depiction of Boceprevir NS3 docked complex using LIGPLOT. Dashed line = hydrogen bond interaction.
Blundell, 1993). Keeping this in view, all inhibitors were docked into protein using Auto Dock/Vina. Docking studies are carried out to study the interactions between active residues and selected inhibitors. The docking scores of top 10 compounds along with structures as well as FDA approved protease inhibitors are provided in Table 2. The lowest binding affinities from Auto Dock/Vina were used as criteria to interpret the best conformation. In order to compare and further validates the interaction mechanism of our selected potent inhibitor, we also compared its docking results with results of FDA approved protease drugs. Among the studied HCV NS5B polymerase inhibitors and reported protease drugs, Compound 1 having binding affinity −14.4 kcal/mol was ranked as most active compound (Table 2). Compound 1 also exhibited a low IC50 value of 0.015, on the basis of which it is considered as second most active compound among the other studied compounds. Compound 1 docked deeply inside the ATP binding site and with few residues of helicase domain. It make closer interactions with Pro228, Ser229, His291, Gln293, Asp294, Arg369, Met413, Thr414,
Thr475, Pro476, Gly477, Glu478, Arg479, Pro480, Ser481, Gly482, Met483, Phe484, Asp485, Ser486, Ala487, Val488, Glu491, Asp494, Ala495, Trp499, Tyr500, Pro521, Val522, Cys523, Gln524, His526, Lys549, Gln550, Gln551, Gly552, Leu553, Asn554, Phe555, Pro556, Asp577, Glu578, Val627 and Thr628 (Fig. 7). The modeled protein, like template, contains an ATP binding site along with residues both from the domains such as helicase as well as protease. High sequence homology was observed as a result of multiple sequence alignment, confirming structural similarity within the family. In order to further validate the binding site of KPB, 3a NS3, its molecular docking study with selected inhibitors was performed. 3.4. Molecular docking Exploring the binding modes is the most important and challenging step for any researcher as it plays the key role in computational chemistry and drug designing. Molecular docking protocols provided insight into the protein interaction and type of inhibition (Sali and 60
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 13. 2D depiction of telaprevir-NS3 docked complex using LIGPLOT.
hydrophobic interactions (Fig. 12). On the other hand, 2D Ligplot analysis of Telaprevir of HCV NS3 for genotype 3a depicted no hydrogen bond interaction, only weak interactions such as hydrophobic interactions were observed (Fig. 13). Additionally, both of these drugs exhibited less binding affinities as compared to our potent inhibitor. Hydrogen bond interactions and binding energy calculations are important factors for investigation of representative docking studies. Hydrogen bond interactions is not only involved in the formation of a stable protein-ligand docked complex but also linked with the alleviation of ligand at target site. H-bond also helps in enhancing the binding affinity and efficacy of drug (Patil et al., 2010). Thus, the suggested NS5B polymerase inhibitor is proven to be better than FDA approved NS3 protease inhibitors. The findings of current study could be helpful in designing more potent dual inhibitors for HCV infection and would be a small contribution in efforts conducted to use NS5B polymerase inhibitors together with NS3.
Ala429, Val430, Glu431, Arg479 and Gln458 (Fig. 8). Examination of protein-ligand docked complex, via VMD, revealed that interactions between modeled protein and best hit for the binding are mainly dominated by hydrogen bond as well as hydrophobic interactions. Six hydrogen bonds between F of ligand and hydrogen atoms of Arg479 (HH21:2.92 Å, HH22:2.99 Å, HH22:2.63 Å, HH12:3.09 Å) and Gln293 (HE22:2.65 Å, HE21:3.91 Å). Hydrophobic contacts between carbon atoms of UNK and carbon atoms of pocket such as His291 (3.99 Å), Gln458 (3.53 Å, 3.67 Å), Pro228 (3.39 Å, 3.90 Å), Thr414 (3.96 Å, 3.97 Å), Met413 (3.99 Å), Asp294 (3.80 Å, 3.70 Å) were observed. Further Ligplot (Wallace et al., 1995) analysis of compound 1 illustrated nine hydrophilic interactions with the key residues of pocket (Fig. 9). As shown in same figure compound 1 formed one hydrogen bond interaction between electronegative atom of UNK and Gln293 at a distance 3.22 Å and play an important role in binding of our suggested potent inhibitor within the cavity of binding site. It is because of all the fore said interactions that the compound 1 showed more potency then telaprevir and boceprevir and other compounds employed in this study. Vigilant assessment of Telaprevir HCV NS3 genotype 3a (−11.7 kcal/mol) as well as Boceprevir-HCV NS3 genotype 3a (−10.4 kcal/mol) docked complexes reveled both compounds occupied the same pocket as like compound 1 (Fig. 10, Fig. 11). Ligplot analysis of Boceprevir for HCV NS3 genotype 3a revealed two hydrogen bonds between Gln 432 and Thr 296 with the electronegative atoms of ligand at distances of 2.82 and 2.90 Å. Rest were
4. Conclusion Computational approaches have been extensively used for the designing of potent small molecules against HCV. Three dimensional structures generated by employing homology modeling technique are extensively used in wide range of applications. In the present study, 3D structure of NS3 genotype 3a, constructed from HCV positive patient, 61
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
was reported for the first time. Keeping in view the importance of protein-inhibitor interactions, we successfully identified a potent inhibitor against the generated model using molecular docking approach. Compound 1 (Table 2) predicted in the current study, showed high binding affinity against modeled protein. It was found binding deep inside the ATP binding site with few residues of helicase domain of NS3. It also developed hydrogen bond interactions unlike FDA approved (Telaprevir and Boceprevir) NS3 protease inhibitors. The study can serve as model to design dual antiviral inhibitors against other HCV genotypes in future.
Ferenci, P., 2006. Pegylated interferon plus ribavirin for chronic hepatitis C: the role of combination therapy today, tomorrow and in the future. Minerva Gastroenterol. Dietol. 52, 157–174. Gu, M., Rice, C.M., 2010. Three conformational snapshots of the hepatitis C virus NS3 helicase reveal a ratchet translocation mechanism. Proc. Natl. Acad. Sci. 107, 521–528. Ibrahim, M., A Saleh, N., M Elshemey, W., A Elsayed, A., 2012a. Fullerene derivative as anti-HIV protease inhibitor: molecular modeling and QSAR approaches. Mini-Rev. Med. Chem. 12, 447–451. Ibrahim, M., A Saleh, N., M Elshemey, W., A Elsayed, A., 2012b. Hexapeptide functionality of cellulose as NS3 protease inhibitors. Med. Chem. 8, 826–830. Kumar, S., Stecher, G., Tamura, K., 2016. MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol. Biol. Evol. 33, 1870–1874 (msw054). Laskowski, R.A., MacArthur, M.W., Moss, D.S., Thornton, J.M., 1993. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291. Li, Z., Wan, H., Shi, Y., Ouyang, P., 2004. Personal experience with four kinds of chemical structure drawing software: review on ChemDraw, ChemWindow, ISIS/draw, and ChemSketch. J. Chem. Inf. Comput. Sci. 44, 1886–1890. Lin, C., Kwong, A., Perni, R., 2006. Discovery and development of VX-950, a novel, covalent, and reversible inhibitor of hepatitis C virus NS3. 4A serine protease. Infect. Disord. Drug Targets 6, 3–16. López-Labrador, F.-X., 2008. Hepatitis C virus NS3/4A protease inhibitors. Recent Pat. Antiinfect. Drug Discov. 3, 157–167. Major, M.E., Feinstone, S.M., 1997. The molecular virology of hepatitis C. Hepatology 25, 1527–1538. Malcolm, B., Liu, R., Lahser, F., Agrawal, S., Belanger, B., Butkiewicz, N., Chase, R., Gheyas, F., Hart, A., Hesk, D., 2006. SCH 503034, a mechanism-based inhibitor of hepatitis C virus NS3 protease, suppresses polyprotein maturation and enhances the antiviral activity of alpha interferon in replicon cells. Antimicrob. Agents Chemother. 50, 1013–1020. Morris, G.M., Huey, R., Lindstrom, W., Sanner, M.F., Belew, R.K., Goodsell, D.S., Olson, A.J., 2009. AutoDock4 and AutoDockTools4: automated docking with selective receptor flexibility. J. Comput. Chem. 30, 2785–2791. Ni, Z.-J., Wagman, A.S., 2004. Progress and development of small molecule HCV antivirals. Curr. Opin. Drug Discov. Devel. 7, 446–459. Patil, R., Das, S., Stanley, A., Yadav, L., Sudhakar, A., Varma, A.K., 2010. Optimized hydrophobic interactions and hydrogen bonding at the target-ligand interface leads the pathways of drug-designing. PLoS One 5, e12029. Perni, R.B., Almquist, S.J., Byrn, R.A., Chandorkar, G., Chaturvedi, P.R., Courtney, L.F., Decker, C.J., Dinehart, K., Gates, C.A., Harbeson, S.L., 2006. Preclinical profile of VX950, a potent, selective, and orally bioavailable inhibitor of hepatitis C virus NS3-4A serine protease. Antimicrob. Agents Chemother. 50, 899–909. Pettersen, E.F., Goddard, T.D., Huang, C.C., Couch, G.S., Greenblatt, D.M., Meng, E.C., Ferrin, T.E., 2004. UCSF chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605–1612. Piccininni, S., Varaklioti, A., Nardelli, M., Dave, B., Raney, K.D., McCarthy, J.E., 2002. Modulation of the hepatitis C virus RNA-dependent RNA polymerase activity by the non-structural (NS) 3 helicase and the NS4B membrane protein. J. Biol. Chem. 277, 45670–45679. Reichard, O., Norkrans, G., Frydén, A., Braconier, J.-H., Sönnerborg, A., Weiland, O., Group, S.S, 1998. Randomised, double-blind, placebo-controlled trial of interferon α2b with and without ribavirin for chronic hepatitis C. Lancet 351, 83–87. Saalau-Bethell, S.M., Woodhead, A.J., Chessari, G., Carr, M.G., Coyle, J., Graham, B., Hiscock, S.D., Murray, C.W., Pathuri, P., Rich, S.J., 2012. Discovery of an allosteric mechanism for the regulation of HCV NS3 protein function. Nat. Chem. Biol. 8, 920–925. Sabri, S., Idrees, M., Rafique, S., Ali, A., Iqbal, M., 2014. Studies on the role of NS3 and NS5A non-structural genes of hepatitis C virus genotype 3a local isolates in apoptosis. Int. J. Infect. Dis. 25, 38–44. Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406–425. Saleh, N.A., Elfiky, A.A., Ezat, A.A., Elshemey, W.M., Ibrahim, M., 2014. The electronic and quantitative structure activity relationship properties of modified telaprevir compounds as HCV NS3 protease inhibitors. J. Comput. Theor. Nanosci. 11, 544–548. Sali, A., Blundell, T., 1993. Comparative protein modeling by satisfaction of spatial restraints. J. Mol. Biol. 234, 779–815. Šali, A., Potterton, L., Yuan, F., van Vlijmen, H., Karplus, M., 1995. Evaluation of comparative protein modeling by MODELLER. Proteins: Struct., Funct., Bioinf. 23, 318–326. Sievers, F., Wilm, A., Dineen, D., Gibson, T.J., Karplus, K., Li, W., Lopez, R., McWilliam, H., Remmert, M., Söding, J., 2011. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol. 7, 539. Tamura, K., Nei, M., Kumar, S., 2004. Prospects for inferring very large phylogenies by using the neighbor-joining method. Proc. Natl. Acad. Sci. U. S. A. 101, 11030–11035. Waheed, Y., Shafi, T., Safi, S.Z., Qadri, I., 2009. Hepatitis C virus in Pakistan: a systematic review of prevalence, genotypes and risk factors. World J. Gastroenterol. 15, 5647–5653. Wallace, A.C., Laskowski, R.A., Thornton, J.M., 1995. LIGPLOT: a program to generate schematic diagrams of protein-ligand interactions. Protein Eng. 8, 127–134. Wasley, A., Alter, M.J., 2000. Epidemiology of Hepatitis C: Geographic Differences and Temporal Trends, Seminars in Liver Disease. Copyright© 2000 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA, pp. 0001–0016 Tel.:+ 1 (212) 584-4663. Wiederstein, M., Sippl, M.J., 2007. ProSA-web: interactive web service for the recognition of errors in three-dimensional structures of proteins. Nucleic Acids Res. 35, W407–W410.
Acknowledgments Authors are thankful to the participants of the study for their cooperation. Financial support from Higher Education Commission (HEC) of Pakistan is highly acknowledged. AZ is supported by indigenous PhD fellowship from HEC. Indigenous PhD Fellowship for 5k Scholars batch 2 PIN NO. 2BM1-566. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.meegid.2018.01.026. References Ali, M., Afzal, S., Zia, A., Hassan, A., Khalil, A.T., Ovais, M., Shinwari, Z.K., Idrees, M., 2016. A systematic review of treatment response rates in Pakistani hepatitis C virus patients; current prospects and future challenges. Medicine 95, e5327. Altschul, S.F., Madden, T.L., Schäffer, A.A., Zhang, J., Zhang, Z., Miller, W., Lipman, D.J., 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402. Ashfaq, U.A., Khan, S.N., Nawaz, Z., Riazuddin, S., 2011. In-vitro model systems to study hepatitis C virus. In: Genetic Vaccines and Therapy. 9. pp. 7. Azam, S.S., Hofer, T.S., Bhattacharjee, A., Lim, L.H.V., Pribil, A.B., Randolf, B.R., Rode, B.M., 2009a. Beryllium (II): the strongest structure-forming ion in water? A QMCF MD simulation study. J. Phys. Chem. B 113, 9289–9295. Azam, S.S., Hofer, T.S., Randolf, B.R., Rode, B.M., 2009b. Germanium (II) in water: an unusual hydration structure results of a QMCF MD simulation. Chem. Phys. Lett. 470, 85–89. Azam, S.S., Hofer, T.S., Randolf, B.R., Rode, B.M., 2009c. Hydration of sodium (I) and potassium (I) revisited: a comparative QM/MM and QMCF MD simulation study of weakly hydrated ions. J. Phys. Chem. A 113, 1827–1834. Azam, S.S., Lim, L.H.V., Hofer, T.S., Randolf, B.R., Rode, B.M., 2010. Hydrated germanium (II): irregular structural and dynamical properties revealed by a quantum mechanical charge field molecular dynamics study. J. Comput. Chem. 31, 278–285. Azam, S.S., Uddin, R., Syed, A.A.S., 2012. Molecular docking studies of potent inhibitors of tyrosinase and α-glucosidase. Med. Chem. Res. 21, 1677–1683. Azam, S.S., Abbasi, S.W., Batool, M., 2014. Structure modeling and docking study of HCV NS5B-3a RNA polymerase for the identification of potent inhibitors. Med. Chem. Res. 23, 618–627. Barreca, M., Carotti, A., Carrieri, A., Chimirri, A., Monforte, A., Calace, M.P., Rao, A., 1999. Comparative molecular field analysis (CoMFA) and docking studies of nonnucleoside HIV-1 RT inhibitors (NNIs). Bioorg. Med. Chem. 7, 2283–2292. Bastos, J.C.S., Padilla, M.A., Caserta, L.C., Miotto, N., Vigani, A.G., Arns, C.W., 2016. Hepatitis C virus: promising discoveries and new treatments. World J. Gastroenterol. 22, 6393. Cheng, F., Shen, J., Luo, X., Zhu, W., Gu, J., Ji, R., Jiang, H., Chen, K., 2002. Molecular docking and 3-D-QSAR studies on the possible antimalarial mechanism of artemisinin analogues. Bioorg. Med. Chem. 10, 2883–2891. Choo, Q.-L., Kuo, G., Weiner, A.J., Overby, L.R., Bradley, D.W., Houghton, M., 1989. Isolation of a cDNA clone derived from a blood-borne non-a, non-B viral hepititis genome. Science 244, 359. Colovos, C., Yeates, T.O., 1993. Verification of protein structures: patterns of nonbonded atomic interactions. Protein Sci. 2, 1511–1519. Dillon, J., 2004. What is the best treatment? J. Viral Hepat. 11, 23–27. Dundas, J., Ouyang, Z., Tseng, J., Binkowski, A., Turpaz, Y., Liang, J., 2006. CASTp: computed atlas of surface topography of proteins with structural and topographical mapping of functionally annotated residues. Nucleic Acids Res. 34, W116–W118. Eisenberg, D., Lüthy, R., Bowie, J.U., 1997. [20] VERIFY3D: assessment of protein models with three-dimensional profiles. Methods Enzymol. 277, 396–404. Elfiky, A.A., Elshemey, W.M., Gawad, W.A., Desoky, O.S., 2013. Molecular modeling comparison of the performance of NS5b polymerase inhibitor (PSI-7977) on prevalent HCV genotypes. Protein J. 32, 75–80. Farci, P., Quinti, I., Farci, S., Alter, H.J., Strazzera, R., Palomba, E., Coiana, A., Cao, D., Casadei, A.M., Ledda, R., 2006. Evolution of hepatitis C viral quasispecies and hepatic injury in perinatally infected children followed prospectively. Proc. Natl. Acad. Sci. 103, 8475–8480.
62
Contents lists available at ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Research paper
Phylogenetic analysis, structure modeling and docking study of HCV NS3 protease for the identification of potent inhibitors Asad Ziaa, Sumra Wajid Abbasib, Shabeer Ahmadd, Muhammad Ziaa, Abida Razac,
T
⁎
a
Department of Biotechnology, Quaid-i-Azam University, Islamabad, Pakistan Department of Bioinformatics, Quaid-i-Azam University, Islamabad, Pakistan c National Institute of Lasers and Optronics, Islamabad, Pakistan d Department of Microbiology and Biotechnology, Abasyn University Peshawar, Pakistan b
A R T I C L E I N F O
A B S T R A C T
Keywords: NS3 helicase Sequencing NS5B polymerase inhibitors Homology modeling Molecular docking
The nonstructural protein 3 (NS3) helicase of HCV is believed to be a plausible target for the identification and designing of potent antiviral drugs. NS3 protein is involved in a positive sense single-stranded viral replication as well as it also cleaves viral poly protein into diverse mature proteins at different time spans. Structural exploration of NS3 revealed that HCV helicase could also act as translocase. In order to identify potential inhibitors for HCV-3a, the current study has been designed. Serum samples from the Pakistani HCV positive patients were collected, sequenced and after purification included in the present study. Phylogenetic analysis on the samples clustered around it in the same group with those from India. Using homology modeling technique, we determined 3D structure of NS3 gene of HCV-3a and employed further in docking studies to discover potent inhibitor against it. As a result of docking Compound 1, with IC50 value of 0.015 and −14.4 kcal/mol energy, ranked as a most pungent inhibitor among all the studied inhibitors. Compound 1 also exhibited good hydrogen bond interactions with the modeled protein. The finding of present study could be used as a lead in future to design an effective dual inhibitor against HCV-3a.
1. Introduction Hepatitis C virus (HCV) belongs to Flaviviridae, is a blood-borne pathogen which infects about 180 million individuals throughout the world; most of the cases lead to end stage liver diseases, fibrosis, cirrhosis and hepatocellular carcinoma (Choo et al., 1989; Major and Feinstone, 1997; Wasley and Alter, 2000). About 10 million populations of Pakistan are infected with HCV, with rare cases of spontaneous clearness; most of them progress to chronic cases (Ali et al., 2016). The most prevalent genotype among eleven HCV genotypes in Pakistan is 3a (Waheed et al., 2009). For a decade pegylated interferon a (PEG-IFN-a) plus ribavirin (RBV) remained treatment of the choice, is expensive, associated with severe side effects and effective for certain genotypes (Dillon, 2004; Farci et al., 2006; Ferenci, 2006; Ni and Wagman, 2004; Reichard et al., 1998). Different treatment response pattern to HCV genotypes is due to the genetic heterogeneity of virus. Considerable variability is shown by the virus which directly interferes with the disease treatment. The response to treatment varies according to HCV genotype and subtype (Bastos et al., 2016). The sustained virological response rate (SVR) in Pakistani population infected with 3a genotype to IFN-α and RBV combination therapy is 87.5%, approximately 2.45% ⁎
of the patients discontinued treatment due to adverse side effects (Ali et al., 2016). Lacking of efficient treatment regimens and increased incidence rate of HCV infection has created a pressure for the therapeutic compounds that can efficiently target the HCV (López-Labrador, 2008). Nonstructural protein NS3/NS4A serine protease and helicase are considered as potential drug targets for the development of effective anti-HCV compounds (Ashfaq et al., 2011). The main role of NS3/NS4A is to cleave viral poly protein into different mature proteins at various time intervals as well as involved in viral replication; HCV helicase affects the viral life cycle at two steps for unwinding of double strand RNA intermediate required for the movement of HCV NS5B polymerase (Piccininni et al., 2002). NS3 structural analysis revealed a new function of HCV helicase as translocase and considered as a potential specific inhibitor to block NS3 helicase (Gu and Rice, 2010). For predication and comparison of molecular and physicochemical properties and mechanisms of reactions of different therapeutic compounds, molecular modeling techniques are widely used (Elfiky et al., 2013; Saleh et al., 2014). For HIV-1 and HCV proteases effective inhibitors are designed through these techniques (Elfiky et al., 2013; Ibrahim et al., 2012a; Ibrahim et al., 2012b). Computational studies not only motivated researchers for the identification of novel therapeutic targets but also
Corresponding author at: Nanotheranostics Research Group, National Institute of Lasers & Optronics (NILOP), Islamabad, Pakistan. E-mail address: [email protected] (A. Raza).
https://doi.org/10.1016/j.meegid.2018.01.026 Received 9 November 2017; Received in revised form 24 January 2018; Accepted 28 January 2018 Available online 31 January 2018 1567-1348/ © 2018 Elsevier B.V. All rights reserved.
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
fortify the drug designing/development process. The field of computational biology and its latest uses has offered reliable methods for the enhanced understanding of biological systems of interest (Azam et al., 2009a; Azam et al., 2009b, 2009c; Azam et al., 2010; Azam et al., 2012; Barreca et al., 1999; Cheng et al., 2002). In the current study, the potent NS3-3a protease inhibitors, previously designed for NS3, have been identified by employing different computational techniques such as homology modeling, and molecular docking. 2. Methods 2.1. Complementary DNA synthesis and amplification of NS3 gene HCV positive patients with genotype 3a from the underprivileged part of Islamabad were collected and included in the present study. RNA was extracted from the serum samples using Virus RNA isolation Kit (INSTANT, AJ Roboscreen, GmbH Germany). The RNA was reverse transcribed in to complimentary DNA (cDNA) using 100 units of M-MLV reversed transcriptase (Invitrogen), with 5 pM of outer antisense primer 5-GGCGACACTCCACCATAGAT-3. Amplification of NS3 gene using sense primer 5-GGCCGTGAGGTGTTGTTGG-3 and anti-sense primer 5TGGTTACTTCCAGATCGGCTG-3 was carried out according to Sabri and colleagues (Sabri et al., 2014). After amplification and purification of the PCR product sequencing was carried out by Sanger method. 2.2. Phylogenetic analysis The sequences obtained from Sanger method were subjected to a phylogenetic analysis for determination of regional distribution of NS3. HCV NS3 sequences from the regions of UK, Germany, France, Australia, India, Nepal, Canada and Somalia were extracted from NCBI database. Alignment and phylogenetic tree construction has been carried out using MEGA 7.0. For multiple sequence alignment clustalw algorithm was used (Sievers et al., 2011). The tree was constructed using neighbor joining algorithm (Fig. 1). Neighbor-Joining method was used in order to infer the evolutionary history (Saitou and Nei, 1987). The optimal tree along with branch length sum equal to 1.88376007 is presented. The phylogenetic tree was drawn to scale in the same branch length units as of evolutionary distance. Maximum Composite (Tamura et al., 2004) Likelihood method was employed to compute evolutionary distances in the number of base substitution per site units. Total sequences involved in the analysis amounted to 32. Missing data and all the positions presenting gaps we eliminated. Final dataset comprised of a total of 1850 positions. Further, MEGA7 was used to conduct evolutionary analysis (Kumar et al., 2016). Fig. 1. 2.3. Homology modeling The crystallographic three-dimensional structure of HCV NS3 genotype 3a, the genotype most prevalent in Pakistani community, has not been reported yet. The sample used in the present study was collected from the underprivileged part of Islamabad, after sequencing the sample was designated KP8/NS3 sequence, in FASTA format. In order to identify suitable templates from RCSB PDB, NCBI BLASTp search of a target sequence was accomplished (http://www.pdb.org/pdb/) (Altschul et al., 1997). BLAST searching resulted in the most appropriate template (PBDID 4B6E) with a resolution of 2.64 Å from Hepatitis C virus (Saalau-Bethell et al., 2012). To generate a 3D model by alignment by means of Clustal Omega, coordinates of 4B6E were used as a template (http://www.ebi.ac.uk/Tools/clustalw2/index.html) (Sievers et al., 2011). To characterize the areas of similarity, pair-wise sequence alignment was performed. This characterization might emphasize evolutionary, functional, structural relationships among target and template two. In total, 5 homology models of NS3 for genotype 3a, with differing geometric conformations, were generated by using MODELLER 9v12
Fig. 1. Phylogenetic analysis of the HCV isolates from the underprivileged part of Islamabad representing the regional distribution of these isolates.
52
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 2. Alignment of HCV NS3 sequence and template 4BE6.Sterics are indicating the conserved residues.
Table 1 PROCHECK, ERRAT, VERIFY_3D and Z-scores for models generated by MODELL. Models
PROCHECK
MODELLER
Residues in core Region (%)
Additionally allowed Region (%)
Generously allowed Region (%)
Disallowed Region (%)
Bad contacts
Model1
481 92.1% 485 92.9% 484 92.7% 479 91.8% 486 93.1%
37 7.1% 33 6.3% 34 6.5% 37 7.1% 31 5.9%
2 0.4% 3 0.6% 3 0.6% 4 0.8% 3 0.6%
2 0.4% 1 0.2% 1 0.2% 2 0.4% 4 0.4%
−0.4 inside −0.3 inside −0.7 inside −0.3 inside −0.4 inside
Model2 Model3 Model4 Model5
ERRAT
Verify_ 3D
ProSA-web
G factor
Quality factor
3D-1D score > 0.2 (%)
Z-score
1.2 better 1.1 better 1.2 better 1.1 better 1.2 better
81.06
93.32%
−8.6
81.09
93.00%
−8.7
77.43
92.21%
−8.5
81.52
91.10%
−8.6
80.51
91.73%
−8.7
score affirms the reliability and validity of studied model. Keeping in mind the end goal to check similarity of modeled protein by arrangement of its own amino acids, Verify3D was utilized. A direct correlation between the value of Verify3D and quality of model exists, higher the value, more better will be the quality. Root mean square deviation (RMSD) values were calculated for the verification of the symmetry of template and target proteins.
(Šali et al., 1995). Among the generated models, the best model was chosen on the basis of PROCHECK analysis (Laskowski et al., 1993). The selected model was then subjected to energy minimization using UCSF Chimera (Pettersen et al., 2004). Gasteiger–Huckel charges were also assigned to protein. A structural evaluation of minimized structure was carried out by using ProSA-web (Wiederstein and Sippl, 2007), Verify3D (Eisenberg et al., 1997) and ERRATv2.0 (Colovos and Yeates, 1993). The stereochemistry of crude model was checked using Ramachandran plot, which is part of PROCHECK program. Energy profile of selected model was obtained by ProSA test as Z-score. Negative Z-
53
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 5. Z-score analysis of Model 5 using ProSa web. Fig. 3. Homology model of HCV NS3 (Pakistani strain 3a) generated using MODELLER.
approved protease inhibitors such as telaprevir and boceprevir were also used (Lin et al., 2006; Malcolm et al., 2006; Perni et al., 2006). Chem Office 2004 was used to draw the bi-dimensional structure of inhibitors. The two dimensional structures of inhibitors were drawn using Chem Office 2004 (Li et al., 2004). Chimera was employed to reduce the energies of selected inhibitors and further used for docking.
2.4. Active site prediction CastP online server was utilized for the detection of active site in the predicted protein model (Dundas et al., 2006). In order to identify the conserved residues the sequence of modeled protein was aligned with the template's sequence (4B6E) using Clustal omega Fig. 2.
2.6. Molecular docking protocol 2.5. HCV inhibitors Graphical User Interface program Auto Dock Tool (ADT) was utilized for the preparatory steps, for instance preparation of pdbqt files for ligands and protein as well as grid box creation (Morris et al., 2009). Auto Dock assigned united atom Kollman charges, polar hydrogen and fragmental volumes to the protein with solvation parameters. The
Keeping in view the possible application of NS5B polymerase inhibitors in combination with NS3 proteases inhibitors to treat HCV, we utilized di-inhibitors reported active against NS5B 3a by Azam and coworkers in the current study (Azam et al., 2014). Additionally, two FDA
Fig. 4. ERRAT plot of best Model 5.
54
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 6. Superimposed 3D structure of target (NS3) and template (4B6E). Target is shown in dim grey whereas template is shown in purple. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 7. CastP predicted active site for HCV NS3. The active site region has been highlighted in green. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
prepared file was saved by ADT in pdbqt format. The grid map was prepared using a grid box utilizing Auto Grid. The grid size was set to 58 × 66 × 48 xyz points with grid spacing of 0.375 Å while at dimensions (x, y, and z): 4.752, 11.341 and −36.749, grid center was placed. The computational time was limited by ascertaining the scoring grid from the ligand structure. Utilizing the protein and ligand information, Auto Dock-Vina was used for docking along with grid box properties. During docking, both protein and ligands were assumed as rigid. The results < 1.0 Å were clustered together in positional root-mean square deviation (RMSD), and represented with the most favorable free energy binding. The pose with lowest energy of binding or affinity was extracted and further aligned for further analysis with the studied protein.
across the world and 4 sequences were used from samples of HCV extracted from patient of underprivileged part of Islamabad, Pakistan. Clustering of these sequences represent the geographical barriers between these regions. Our sample sequence were found to be clustering with the Indo-Pak peninsula while the rest of sequences where grouped according to continents. Only one of the UK sequence showed abnormality in clustering which may have arisen by constant travelling of individual from Indo-Pak peninsula to Great Britain. From our analysis we can confirm that these are specific to Indo-Pak peninsula and further drug therapy on these sequences would be more beneficial than a generalized therapy around the world.
3.2. Homology modeling and model validation 3. Results and discussion The availability of 3D structure is very crucial in order to explore the potent inhibitors and to study their interactions (Sali and Blundell, 1993). The structure of NS3 of HCV genotype 3a has not been predicted yet, so it created a need for the development of its homology model.
3.1. Phylogenetic analysis Analysis was carried out using 28 different regional sequences from 55
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Table 2 Binding affinities and IC50 values of Top 10 active compounds and FDA approved NS3 protease inhibitors against NS3. S. No.
Compounds
1
IC50 (μM)
Binding affinities
0.015
−14.4
0.014
−13.5
0.075
−13.4
0.024
−13.3
0.019
−13.1
0.038
−12.7
17 of 3-Heterocyclylquinolone
2 16 of 3-Heterocyclylquinolone
3 14 of 3-Heterocyclylquinolone
4 15 of 3-Heterocyclylquinolone
5 13 of 3-Heterocyclylquinolone
6 18 of 3-Heterocyclylquinolone
(continued on next page)
56
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Table 2 (continued) S. No.
Compounds
IC50 (μM)
Binding affinities
1.6
−12.5
0.023
−12.3
Telaprevir
< 0.2
−11.7
Boceprevir
0.5
−10.4
7 7 of 3-Heterocyclylquinolone
8 11 of 3-Heterocyclylquinolone
indicating the good quality and suitability of the generated model. In the next step, to further optimize the Model5energy minimization of was accomplished. The minimized structure of the best model is provided in Fig. 3. In order to evaluate the compatibility of generated model and amino acid sequence, the best model was assessed by means of various verification web tools such as ERRAT, ProSA and Verify3D. The overall quality factor estimated via ERRAT was 80.51% for KP8, 3a NS3, showing the good quality of structure (Fig. 4, Table 1). The Z-score of −8.7, for the best model, was obtained using ProSA web server (Fig. 5, Table 1). Lastly, Verify3D was used to calculate the packing quality of every amino acid. By making use of a scoring function, it made a comparison between residues and their environment in generated models and assigned a 3D–1D score (> 0.2) for each and every modeled protein. In current case 91.73% of the residues of Model 5 occupied an averaged 3D-1D score greater than and equal to 2 (Table 1). Thus, affirming the
The template selection is thought to be the most important step in homology modeling as the accuracy of model depends on it. NS3/4a protein from Hepatitis C virus (PDB entry: 4B6E) was selected as a template, with 100% of sequence coverage, 78% identity and e-value 0.0. Higher similarity between template and target might be a plausible reason for the similar geometric arrangement. So sequence alignment confirmed the protein conservation on sequence level. Homology models for query sequence were predicted using protein modeling tool MODELLER. MODELLER generated 5 models among which the best model was selected on the basis of Ramachandran Plot, generated by PROCHECK. A comparison of models generated by MODELLER, based on overall stereochemistry, illuminated that model 5was best among all. Model 5exhibited good percent of allowed, disallowed and general region in comparison with other models in Table 1. Ramachandran plot of the Model5 showed that 93.1% of the residues lied in the most favored regions, whereas 0.4% of the residues lied in the disallowed regions. Overall G factor value, 1.2, was also in acceptable range 57
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 8. 3D representation of compund1-NS3 docked complex generated using Chimera. The inset represents magnified view of binding pocket. Fig. 9. 2D depiction of compund1-NS3 docked complex using LIGPLOT. Dashed line = hydrogen bond interaction.
58
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 10. 3D representation of Telaprevir NS3 docked complex generated using Chimera. The inset represent magnified view of binding pocket.
Fig. 11. 3D representation of Boceprevir NS3 docked complex generated using Chimera. The inset represent magnified view of binding pocket.
Gly29, Glu30, Val31, Gln32, Val33, Leu34, Ser35, Thr36, Ala37, Thr38, Gln39, Thr40, Phe41, Leu42, Thr45, Met50, Tyr54, His55, Gly56, Ala57, Gly58, Arg60, Thr61, Leu62, Ala63, Gly64, Val65, Lys66, Pro68, Ala69, Val76, Asp79, Trp83, Pro84, Ala85, Pro86, Pro87, Gly88, Ala89, Leu92, Val105, Thr106, Arg107, Ala109, Lys134, Pro140, Met142, Arg153, Pro203, Thr204, Gly205, Gly207, Lys208, Ser209, Thr210, Asn227, Pro228, Ser209, Val230, Ala231, Ala232, Gly235, Phe236, Ser238, Phe239, Arg242, Gly253, Asn254, Thr267, Tyr268, Gly269, Lys270, Asp288, Glu289, His291, Ala292, Gln293, Asp294, Ser295, Thr296, Ser297, Ala321, Thr322, Pro323, Pro324, His331, Asn333, Leu334, Ile367, Arg369, Lys370, Met373, Tyr389, Tyr390, Arg391, Gly392, Leu393, Asp394, Val395, Ile398, Thr400, Thr409, Asp410, Ala411, Met413, Thr414, Gly415, Phe416, Thr417, Gly418, Ala429, Val430, Glu431, Gln432, Tyr433, Val434, Phe436, Glu445, Thr446, Arg447, Thr448, Ala449, Pro450, Asp452, Val454, Ser455, Ser457, Gln458, Arg459, Arg460, Gly461, Arg462, Thr463, Gly464, Arg465,
accuracy and compatibility of selected model with its sequence. Further, the topology of the target and the template protein was verified by superimposing both the structures using Chimera which gives the root mean square deviation (RMSD) value of 0.59 Å (6).When the structure of minimized protein was closely observed, it depicted the same structural topology as of the template. Template protein (4B6E) consists of two domains: protease domain and helicase domain. The concurrence of the said domains in a protein has a plays a vital role maintaining the affinity, selectivity and catalytic rate for their particular substrates (Saalau-Bethell et al., 2012) (Fig. 6). 3.3. Active site prediction The important active site residues identified by Cast Pare Ile1, Thr2, Ala3, Tyr4, Ala5, Gln6, Gln7, Thr8, Arg9, Arg10, Leu11, Thr14, Ile15, Thr17, Ser18, Leu19, Gly21, Asp23, Asn 25, Val26, Val27, Ala28, 59
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 12. 2D depiction of Boceprevir NS3 docked complex using LIGPLOT. Dashed line = hydrogen bond interaction.
Blundell, 1993). Keeping this in view, all inhibitors were docked into protein using Auto Dock/Vina. Docking studies are carried out to study the interactions between active residues and selected inhibitors. The docking scores of top 10 compounds along with structures as well as FDA approved protease inhibitors are provided in Table 2. The lowest binding affinities from Auto Dock/Vina were used as criteria to interpret the best conformation. In order to compare and further validates the interaction mechanism of our selected potent inhibitor, we also compared its docking results with results of FDA approved protease drugs. Among the studied HCV NS5B polymerase inhibitors and reported protease drugs, Compound 1 having binding affinity −14.4 kcal/mol was ranked as most active compound (Table 2). Compound 1 also exhibited a low IC50 value of 0.015, on the basis of which it is considered as second most active compound among the other studied compounds. Compound 1 docked deeply inside the ATP binding site and with few residues of helicase domain. It make closer interactions with Pro228, Ser229, His291, Gln293, Asp294, Arg369, Met413, Thr414,
Thr475, Pro476, Gly477, Glu478, Arg479, Pro480, Ser481, Gly482, Met483, Phe484, Asp485, Ser486, Ala487, Val488, Glu491, Asp494, Ala495, Trp499, Tyr500, Pro521, Val522, Cys523, Gln524, His526, Lys549, Gln550, Gln551, Gly552, Leu553, Asn554, Phe555, Pro556, Asp577, Glu578, Val627 and Thr628 (Fig. 7). The modeled protein, like template, contains an ATP binding site along with residues both from the domains such as helicase as well as protease. High sequence homology was observed as a result of multiple sequence alignment, confirming structural similarity within the family. In order to further validate the binding site of KPB, 3a NS3, its molecular docking study with selected inhibitors was performed. 3.4. Molecular docking Exploring the binding modes is the most important and challenging step for any researcher as it plays the key role in computational chemistry and drug designing. Molecular docking protocols provided insight into the protein interaction and type of inhibition (Sali and 60
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
Fig. 13. 2D depiction of telaprevir-NS3 docked complex using LIGPLOT.
hydrophobic interactions (Fig. 12). On the other hand, 2D Ligplot analysis of Telaprevir of HCV NS3 for genotype 3a depicted no hydrogen bond interaction, only weak interactions such as hydrophobic interactions were observed (Fig. 13). Additionally, both of these drugs exhibited less binding affinities as compared to our potent inhibitor. Hydrogen bond interactions and binding energy calculations are important factors for investigation of representative docking studies. Hydrogen bond interactions is not only involved in the formation of a stable protein-ligand docked complex but also linked with the alleviation of ligand at target site. H-bond also helps in enhancing the binding affinity and efficacy of drug (Patil et al., 2010). Thus, the suggested NS5B polymerase inhibitor is proven to be better than FDA approved NS3 protease inhibitors. The findings of current study could be helpful in designing more potent dual inhibitors for HCV infection and would be a small contribution in efforts conducted to use NS5B polymerase inhibitors together with NS3.
Ala429, Val430, Glu431, Arg479 and Gln458 (Fig. 8). Examination of protein-ligand docked complex, via VMD, revealed that interactions between modeled protein and best hit for the binding are mainly dominated by hydrogen bond as well as hydrophobic interactions. Six hydrogen bonds between F of ligand and hydrogen atoms of Arg479 (HH21:2.92 Å, HH22:2.99 Å, HH22:2.63 Å, HH12:3.09 Å) and Gln293 (HE22:2.65 Å, HE21:3.91 Å). Hydrophobic contacts between carbon atoms of UNK and carbon atoms of pocket such as His291 (3.99 Å), Gln458 (3.53 Å, 3.67 Å), Pro228 (3.39 Å, 3.90 Å), Thr414 (3.96 Å, 3.97 Å), Met413 (3.99 Å), Asp294 (3.80 Å, 3.70 Å) were observed. Further Ligplot (Wallace et al., 1995) analysis of compound 1 illustrated nine hydrophilic interactions with the key residues of pocket (Fig. 9). As shown in same figure compound 1 formed one hydrogen bond interaction between electronegative atom of UNK and Gln293 at a distance 3.22 Å and play an important role in binding of our suggested potent inhibitor within the cavity of binding site. It is because of all the fore said interactions that the compound 1 showed more potency then telaprevir and boceprevir and other compounds employed in this study. Vigilant assessment of Telaprevir HCV NS3 genotype 3a (−11.7 kcal/mol) as well as Boceprevir-HCV NS3 genotype 3a (−10.4 kcal/mol) docked complexes reveled both compounds occupied the same pocket as like compound 1 (Fig. 10, Fig. 11). Ligplot analysis of Boceprevir for HCV NS3 genotype 3a revealed two hydrogen bonds between Gln 432 and Thr 296 with the electronegative atoms of ligand at distances of 2.82 and 2.90 Å. Rest were
4. Conclusion Computational approaches have been extensively used for the designing of potent small molecules against HCV. Three dimensional structures generated by employing homology modeling technique are extensively used in wide range of applications. In the present study, 3D structure of NS3 genotype 3a, constructed from HCV positive patient, 61
Infection, Genetics and Evolution 59 (2018) 51–62
A. Zia et al.
was reported for the first time. Keeping in view the importance of protein-inhibitor interactions, we successfully identified a potent inhibitor against the generated model using molecular docking approach. Compound 1 (Table 2) predicted in the current study, showed high binding affinity against modeled protein. It was found binding deep inside the ATP binding site with few residues of helicase domain of NS3. It also developed hydrogen bond interactions unlike FDA approved (Telaprevir and Boceprevir) NS3 protease inhibitors. The study can serve as model to design dual antiviral inhibitors against other HCV genotypes in future.
Ferenci, P., 2006. Pegylated interferon plus ribavirin for chronic hepatitis C: the role of combination therapy today, tomorrow and in the future. Minerva Gastroenterol. Dietol. 52, 157–174. Gu, M., Rice, C.M., 2010. Three conformational snapshots of the hepatitis C virus NS3 helicase reveal a ratchet translocation mechanism. Proc. Natl. Acad. Sci. 107, 521–528. Ibrahim, M., A Saleh, N., M Elshemey, W., A Elsayed, A., 2012a. Fullerene derivative as anti-HIV protease inhibitor: molecular modeling and QSAR approaches. Mini-Rev. Med. Chem. 12, 447–451. Ibrahim, M., A Saleh, N., M Elshemey, W., A Elsayed, A., 2012b. Hexapeptide functionality of cellulose as NS3 protease inhibitors. Med. Chem. 8, 826–830. Kumar, S., Stecher, G., Tamura, K., 2016. MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol. Biol. Evol. 33, 1870–1874 (msw054). Laskowski, R.A., MacArthur, M.W., Moss, D.S., Thornton, J.M., 1993. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291. Li, Z., Wan, H., Shi, Y., Ouyang, P., 2004. Personal experience with four kinds of chemical structure drawing software: review on ChemDraw, ChemWindow, ISIS/draw, and ChemSketch. J. Chem. Inf. Comput. Sci. 44, 1886–1890. Lin, C., Kwong, A., Perni, R., 2006. Discovery and development of VX-950, a novel, covalent, and reversible inhibitor of hepatitis C virus NS3. 4A serine protease. Infect. Disord. Drug Targets 6, 3–16. López-Labrador, F.-X., 2008. Hepatitis C virus NS3/4A protease inhibitors. Recent Pat. Antiinfect. Drug Discov. 3, 157–167. Major, M.E., Feinstone, S.M., 1997. The molecular virology of hepatitis C. Hepatology 25, 1527–1538. Malcolm, B., Liu, R., Lahser, F., Agrawal, S., Belanger, B., Butkiewicz, N., Chase, R., Gheyas, F., Hart, A., Hesk, D., 2006. SCH 503034, a mechanism-based inhibitor of hepatitis C virus NS3 protease, suppresses polyprotein maturation and enhances the antiviral activity of alpha interferon in replicon cells. Antimicrob. Agents Chemother. 50, 1013–1020. Morris, G.M., Huey, R., Lindstrom, W., Sanner, M.F., Belew, R.K., Goodsell, D.S., Olson, A.J., 2009. AutoDock4 and AutoDockTools4: automated docking with selective receptor flexibility. J. Comput. Chem. 30, 2785–2791. Ni, Z.-J., Wagman, A.S., 2004. Progress and development of small molecule HCV antivirals. Curr. Opin. Drug Discov. Devel. 7, 446–459. Patil, R., Das, S., Stanley, A., Yadav, L., Sudhakar, A., Varma, A.K., 2010. Optimized hydrophobic interactions and hydrogen bonding at the target-ligand interface leads the pathways of drug-designing. PLoS One 5, e12029. Perni, R.B., Almquist, S.J., Byrn, R.A., Chandorkar, G., Chaturvedi, P.R., Courtney, L.F., Decker, C.J., Dinehart, K., Gates, C.A., Harbeson, S.L., 2006. Preclinical profile of VX950, a potent, selective, and orally bioavailable inhibitor of hepatitis C virus NS3-4A serine protease. Antimicrob. Agents Chemother. 50, 899–909. Pettersen, E.F., Goddard, T.D., Huang, C.C., Couch, G.S., Greenblatt, D.M., Meng, E.C., Ferrin, T.E., 2004. UCSF chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605–1612. Piccininni, S., Varaklioti, A., Nardelli, M., Dave, B., Raney, K.D., McCarthy, J.E., 2002. Modulation of the hepatitis C virus RNA-dependent RNA polymerase activity by the non-structural (NS) 3 helicase and the NS4B membrane protein. J. Biol. Chem. 277, 45670–45679. Reichard, O., Norkrans, G., Frydén, A., Braconier, J.-H., Sönnerborg, A., Weiland, O., Group, S.S, 1998. Randomised, double-blind, placebo-controlled trial of interferon α2b with and without ribavirin for chronic hepatitis C. Lancet 351, 83–87. Saalau-Bethell, S.M., Woodhead, A.J., Chessari, G., Carr, M.G., Coyle, J., Graham, B., Hiscock, S.D., Murray, C.W., Pathuri, P., Rich, S.J., 2012. Discovery of an allosteric mechanism for the regulation of HCV NS3 protein function. Nat. Chem. Biol. 8, 920–925. Sabri, S., Idrees, M., Rafique, S., Ali, A., Iqbal, M., 2014. Studies on the role of NS3 and NS5A non-structural genes of hepatitis C virus genotype 3a local isolates in apoptosis. Int. J. Infect. Dis. 25, 38–44. Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406–425. Saleh, N.A., Elfiky, A.A., Ezat, A.A., Elshemey, W.M., Ibrahim, M., 2014. The electronic and quantitative structure activity relationship properties of modified telaprevir compounds as HCV NS3 protease inhibitors. J. Comput. Theor. Nanosci. 11, 544–548. Sali, A., Blundell, T., 1993. Comparative protein modeling by satisfaction of spatial restraints. J. Mol. Biol. 234, 779–815. Šali, A., Potterton, L., Yuan, F., van Vlijmen, H., Karplus, M., 1995. Evaluation of comparative protein modeling by MODELLER. Proteins: Struct., Funct., Bioinf. 23, 318–326. Sievers, F., Wilm, A., Dineen, D., Gibson, T.J., Karplus, K., Li, W., Lopez, R., McWilliam, H., Remmert, M., Söding, J., 2011. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol. 7, 539. Tamura, K., Nei, M., Kumar, S., 2004. Prospects for inferring very large phylogenies by using the neighbor-joining method. Proc. Natl. Acad. Sci. U. S. A. 101, 11030–11035. Waheed, Y., Shafi, T., Safi, S.Z., Qadri, I., 2009. Hepatitis C virus in Pakistan: a systematic review of prevalence, genotypes and risk factors. World J. Gastroenterol. 15, 5647–5653. Wallace, A.C., Laskowski, R.A., Thornton, J.M., 1995. LIGPLOT: a program to generate schematic diagrams of protein-ligand interactions. Protein Eng. 8, 127–134. Wasley, A., Alter, M.J., 2000. Epidemiology of Hepatitis C: Geographic Differences and Temporal Trends, Seminars in Liver Disease. Copyright© 2000 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA, pp. 0001–0016 Tel.:+ 1 (212) 584-4663. Wiederstein, M., Sippl, M.J., 2007. ProSA-web: interactive web service for the recognition of errors in three-dimensional structures of proteins. Nucleic Acids Res. 35, W407–W410.
Acknowledgments Authors are thankful to the participants of the study for their cooperation. Financial support from Higher Education Commission (HEC) of Pakistan is highly acknowledged. AZ is supported by indigenous PhD fellowship from HEC. Indigenous PhD Fellowship for 5k Scholars batch 2 PIN NO. 2BM1-566. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.meegid.2018.01.026. References Ali, M., Afzal, S., Zia, A., Hassan, A., Khalil, A.T., Ovais, M., Shinwari, Z.K., Idrees, M., 2016. A systematic review of treatment response rates in Pakistani hepatitis C virus patients; current prospects and future challenges. Medicine 95, e5327. Altschul, S.F., Madden, T.L., Schäffer, A.A., Zhang, J., Zhang, Z., Miller, W., Lipman, D.J., 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402. Ashfaq, U.A., Khan, S.N., Nawaz, Z., Riazuddin, S., 2011. In-vitro model systems to study hepatitis C virus. In: Genetic Vaccines and Therapy. 9. pp. 7. Azam, S.S., Hofer, T.S., Bhattacharjee, A., Lim, L.H.V., Pribil, A.B., Randolf, B.R., Rode, B.M., 2009a. Beryllium (II): the strongest structure-forming ion in water? A QMCF MD simulation study. J. Phys. Chem. B 113, 9289–9295. Azam, S.S., Hofer, T.S., Randolf, B.R., Rode, B.M., 2009b. Germanium (II) in water: an unusual hydration structure results of a QMCF MD simulation. Chem. Phys. Lett. 470, 85–89. Azam, S.S., Hofer, T.S., Randolf, B.R., Rode, B.M., 2009c. Hydration of sodium (I) and potassium (I) revisited: a comparative QM/MM and QMCF MD simulation study of weakly hydrated ions. J. Phys. Chem. A 113, 1827–1834. Azam, S.S., Lim, L.H.V., Hofer, T.S., Randolf, B.R., Rode, B.M., 2010. Hydrated germanium (II): irregular structural and dynamical properties revealed by a quantum mechanical charge field molecular dynamics study. J. Comput. Chem. 31, 278–285. Azam, S.S., Uddin, R., Syed, A.A.S., 2012. Molecular docking studies of potent inhibitors of tyrosinase and α-glucosidase. Med. Chem. Res. 21, 1677–1683. Azam, S.S., Abbasi, S.W., Batool, M., 2014. Structure modeling and docking study of HCV NS5B-3a RNA polymerase for the identification of potent inhibitors. Med. Chem. Res. 23, 618–627. Barreca, M., Carotti, A., Carrieri, A., Chimirri, A., Monforte, A., Calace, M.P., Rao, A., 1999. Comparative molecular field analysis (CoMFA) and docking studies of nonnucleoside HIV-1 RT inhibitors (NNIs). Bioorg. Med. Chem. 7, 2283–2292. Bastos, J.C.S., Padilla, M.A., Caserta, L.C., Miotto, N., Vigani, A.G., Arns, C.W., 2016. Hepatitis C virus: promising discoveries and new treatments. World J. Gastroenterol. 22, 6393. Cheng, F., Shen, J., Luo, X., Zhu, W., Gu, J., Ji, R., Jiang, H., Chen, K., 2002. Molecular docking and 3-D-QSAR studies on the possible antimalarial mechanism of artemisinin analogues. Bioorg. Med. Chem. 10, 2883–2891. Choo, Q.-L., Kuo, G., Weiner, A.J., Overby, L.R., Bradley, D.W., Houghton, M., 1989. Isolation of a cDNA clone derived from a blood-borne non-a, non-B viral hepititis genome. Science 244, 359. Colovos, C., Yeates, T.O., 1993. Verification of protein structures: patterns of nonbonded atomic interactions. Protein Sci. 2, 1511–1519. Dillon, J., 2004. What is the best treatment? J. Viral Hepat. 11, 23–27. Dundas, J., Ouyang, Z., Tseng, J., Binkowski, A., Turpaz, Y., Liang, J., 2006. CASTp: computed atlas of surface topography of proteins with structural and topographical mapping of functionally annotated residues. Nucleic Acids Res. 34, W116–W118. Eisenberg, D., Lüthy, R., Bowie, J.U., 1997. [20] VERIFY3D: assessment of protein models with three-dimensional profiles. Methods Enzymol. 277, 396–404. Elfiky, A.A., Elshemey, W.M., Gawad, W.A., Desoky, O.S., 2013. Molecular modeling comparison of the performance of NS5b polymerase inhibitor (PSI-7977) on prevalent HCV genotypes. Protein J. 32, 75–80. Farci, P., Quinti, I., Farci, S., Alter, H.J., Strazzera, R., Palomba, E., Coiana, A., Cao, D., Casadei, A.M., Ledda, R., 2006. Evolution of hepatitis C viral quasispecies and hepatic injury in perinatally infected children followed prospectively. Proc. Natl. Acad. Sci. 103, 8475–8480.
62