This article is mentioned in:

Introduction

The viral family Flaviviridae includes the genera Flavivirus, Pestivirus, Pegivirus and Hepacivirus. Arthropod vectors, mainly ticks and mosquitoes, constitute the transmission pathway of Flaviviridae viruses, causing epidemics and medical concerns due to the large number of diseases that they inflict on both humans and animals (1,2). The different members of the Flaviviridae family share some common elements of viral organization. Their viral particles (virions) are small (~50 nm), spherical and enveloped, that incorporate a single-stranded RNA of 9.5-12.5 kb (1). The viral genome is located inside the capsid of the virion, having a positive-sense polarity and a long open reading frame, which is flanked by untranslated regions at the 5' and 3' ends. The translated polyprotein consists of three structural [capsid (C), membrane (M) and envelope (E)] and seven non-structural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B and NS5). The latter region of NS proteins, at the C-terminal part of the polyprotein, has a great contribution in the RNA replication process. The most crucial NS proteins are the viral helicase and the viral RNA-dependent RNA polymerase (RdRp) (2). The viral helicase constitutes the main subject in the present study.

The viral RNA helicases are attributed to the NS3 region of the viral polyprotein, and they are involved in duplex unwinding, during viral RNA replication. Being a promising antiviral target, helicase inhibition leads to the pause of the replication, proliferation and consequently, to the survival and transmission of Flaviviridae viruses (3). Spondweni virus (SPONV) is a member of the Flaviviridae virus family, of the genus Flavivirus, and belongs to a serogroup with the Zika virus. SPONV was first detected in Nigeria and South Africa in the 1950s and subsequently, in sub-Saharan Africa. The viral cycle is located between mosquitoes and non-human primates, causing symptomatic infections of mild illness. However, some cases result in more severe disease, including febrile syndrome, Vascular leakage (shock) or neurological impairments. The epidemic of Zika virus that broke out in the Western Hemisphere, reveals the epidemic potential of SPONV and the increased number of infections in the near future (4,5).

To date, there is no anti-SPONV therapy available, while at the same time, the continued threat of emerging Flavivirus remains incurable. This condition highlights the need for an extensive fundamental study of viral biology, in order to develop novel antiviral vaccines and drugs. The present study proposes the three-dimensional structure of the helicase/protease enzyme of SPONV, since its 3D structure has not yet been resolved. The structure is modelled by applying homology modelling techniques and using the crystal structure of the Dengue virus-4 (DENV-4) helicase/protease (PDB: 2VBC), of the same viral family (Flaviviridae), as the template (6). The next challenge of the study was the confirmation testing of the functionality, efficacy and reliability of the model in structure-based drug design strategies. For this purpose, the resolved structure of the hepatitis C virus (HCV) helicase (PDB: 1A1V) was used, as it is complexed with a single-stranded RNA, a key molecule for the establishment of interactions with a future inhibitor of the SPONV helicase (7).

Materials and methods

Database sequence search

The amino acid sequence data of the NS3 helicase of the Flaviviridae family were collected from the NCBI database, by using related keywords and Virus Pathogen Database and Analysis Resource (ViPR) (8,9). The amino acid sequence of the SPONV NS3 protein was obtained from the NCBI database, with the accession no. YP_009227191.1 and entry name, Spondweni virus, non-structural protein NS3. The sequence length is 619 aa. All the available NS3 viral protein sequences were filtered in order to remove the irrelevant sequences, as well as the hypothetical, partial and synthetic sequences. Moreover, the final dataset of the NS3 viral proteins was merged using the dataset from previous research, as previously described by Papageorgiou et al (1).

Multiple sequence alignment (MSA) and conserved motifs

MSA was executed using the MATLAB Bioinformatics toolbox (https://uk.mathworks.com/products/bioinfo.html), utilizing a guide tree and the progressive MSA method as previously described (10,11). Pairwise distances among sequences were estimated based on the pairwise alignment with the ‘Gonnet’ method and followed by calculating the differences between each pair of sequences (12). The Neighbor-Joining method was used towards to estimating the guide tree by assuming equal variance and independence of evolutionary distance estimates (13). Finally, the major conserved helicase motifs, which are characteristic of the Flaviviridae family, were identified using the consensus sequences from the MSA. The visualization was performed using the Jalview program, providing important information on sequence conservation, quality and consensus (14).

Phylogenetic analysis

The construction of a phylogenetic tree is a consequence of the MSA. For the purposes of the present study, the identification of the homologous viral structure that will be used as a template for homology modeling, a phylogenetic tree was constructed with representatives of the genus Flavivirus. The formation of separate groups among members of the genus Flavivirus helps to identify which virus is the most closely related to SPONV. The method that was used to create a phylogenetic tree was the Neighbor-Joining method and the scores were computed with the scoring matrix BLOSUM62. The visualization of the tree was completed using the Jalview program (14-16).

Template identification

The BLASTp algorithm was used to identify the template structure for the homology modeling procedure of SPONV helicase/protease, by searching the Protein Data Bank (PDB) (17,18). The most important alignments emerged and according to the identity percentage and the crystal resolution of the structures, the structure that will be the template can be distinguished. Subsequently, another search was performed through the Molecular Operating Environment (MOE) program (www.chemcomp.com) against the PDB.

Homology modelling

Homology modelling of the SPONV NS3 helicase was performed using MOE version 2016.0801 and its homology modelling application (19-21). All calculations and visual constructions were performed using this program. The MOE homology model method is separated into four main steps (22). First, a primary fragment geometry specification. Second, the insertion and deletions task. The third step is the loop selection and the side-chain packing and the last step is the final model selection and refinement.

Model evaluation

The evaluation of the quality and reliability of the produced SPONV helicase model is vital for the viability of the present study. As a result, the created model was evaluated within the MOE package. For this purpose, the RMSD and Geometry-Ramachandran Plot diagrams were calculated and their contribution to the structural estimation of the protein's quality is crucial (23,24). The RMSD calculation helps to understand the quality of the system, while high RMSD values indicate poor quality systems as opposed to low values, which are indicative of good quality systems. Moreover, the Ramachandran plot is the most reliable in silico tool for the enzyme's stereochemical evaluation, looking at its phi/psi dihedral angles. Last but not least, in order to analyze the molecular surface of the produced SPONV helicase model, the electrostatic potential surface was calculated by solving the non-linear Poisson-Boltzmann equation using MOE software (20).

Identification of key amino acid residues as candidate targets

A structural analysis was performed using the SPONV model and the structural features from other available 3D structures in an effort to confirm the functionality, suitability and reliability of the SPONV helicase model. The HCV helicase (PDB: 1A1V) was the studied crystal structure, since several structural-based drug design studies have been performed using this crystal structure and they provide beneficial knowledge and a number of pharmacophore models and inhibitors that have been already designed (25-27). Last but not least, some important conserved residues are isolated and selected as a suitable target for structure-based experiments. Based on this strategy, the present study attempted to exploit the existing inhibitor molecules of HCV in order to inhibit the activity of SPONV virus helicase.

Results and Discussion

MSA and phylogenetic analysis

The NS3 domain of Flaviviridae consists of both the protease and the helicase coding regions (1). MSA of the NS3 protein sequences indicate conservation in known and unknown important regions within all Flaviviridae viruses, throughout the whole length of the sequence (2). It is worth mentioning that the NS3 sequences of the genus Pestivirus, include an extended insertion at the N-terminal half of the protein, which differentiates their size from Flavivirus, Hepacivirus and Pegivirus. The conserved regions in the four district genera indicate the several critical functional domains of the enzyme. Based on the results from the extracted consensus sequence of the NS3 MSA, all known conserved regions were identified and highlighted (Fig. 1). Phylogenetic analysis of the representative NS3 viral protein sequences indicates a clear separation between the genera Flavivirus, Hepacivirus, Pestivirus and Pegivirus of the Flaviviridae family (1). In the present study, the phylogenetic tree was performed for the genus Flavivirus, in which SPONV belongs (Fig. 2). The SPONV NS3 viral protein was identified in the same monophyletic branch with Zika virus, and DENV-4.

Figure 1 Consensus sequence, amino acids quality and conservation based on the NS3 multiple sequence alignment results are performed using Jalview program. The seven major conserved motifs of the Flavivirus helicases (black boxes) and the three suggested conserved regions (red boxes) are marked.

Figure 2 Phylogenetic tree using the NS3 non-structural representative protein sequences dataset of Flavivirus.

The BLASTp algorithm was identified several protein structures as a candidate protein templates in order to perform the homology modelling of the NS3 viral protein of the SPONV. Two parameters have been studied for the optimal selection of the final template including i) the sequence identity; and ii) the crystal structure resolution (2,16). The selected template was the crystal structure of the NS3 protease/helicase from DENV-4 (PDB: 2VBC) (6). The DENV-4 belongs to the same viral family (Flaviviridae) and genus (Flavivirus). The 2VBC DENV-4 protease/helicase structure has been established by X-ray crystallography at a 3.15 Å resolution, having an identity percentage of 65,59%, a query cover of 100% and a sequence length of 618 aa (6). Moreover, the crystal structure of the HCV NS3 helicase, which has been co-crystalized with a single-stranded DNA molecule (ssDNA), at a resolution of 2.2 Å (PDB: 1A1V), has provided further information of the functionality of the NS3 viral protein (7). HCV belongs to the same family as SPONV (Flaviviridae) (1). The selection was made due to the presence of ssDNA, contained in the crystal, but also due to the extensive study of the structure for potential pharmacological targets. Furthermore, all the available results and findings, as described in the study by Vlachakis et al (3), for a possible pharmacophore model against the viral NS3 protein of the Classical Swine Fever virus, were studied. Classical swine fever virus is also a member of the genus Flavivirus.

A MSA was constructed, including the SPONV helicase/protease sequence, the DENV-4 helicase/protease sequence (2VBC) and the HCV helicase sequence (1A1V) (Fig. 3). The alignment revealed the seven major conserved motifs, which are characteristic and unique to the helicases of the Flaviviridae viral family (motifs 1-6) and these are highlighted in Fig. 2. Moreover, in the MSA identified certain ‘key’ amino acids that are necessary for the NS3 viral protein inhibition, as described in the study by Vlachakis et al (3).

Figure 3 Sequence alignment between the Spondweni virus helicase/protease sequence, the corresponding sequence of the Dengue virus-4 helicase template (RCSB entry: 2VBC) and the sequence of the hepatitis C virus (RCSB entry: 1A1V). All seven major conserved motifs of Flaviviridae helicases are highlighted as M1-M6. SPONV, Spondweni virus; DENV-4, Dengue virus-4; HCV, hepatitis C virus.

Description of the SPONV helicase model

The candidate model of the SPONV helicase/protease was established using the homology modeling process of the MOE package (Fig. 4). Although the Zika virus belongs to the same antigenic group with SPONV, and consequently they have a similar helicase structure, the selected template belongs to Dengue virus due to a greater query cover with 100% identity in sequence length, a relative higher identity percentage throughout the protein sequence and a good-quality crystal structure resolution (3.15 Å). Summarizing the results from the MSA and the model, in the SPONV helicase/protease model, all the structural features of known Flaviviridae helicases were identified (Figs. 2 and 4). The model shares a similar structural topology to its template (2VBC) (Fig. 4). Particularly, in the SPONV helicase/protease model, the three distinct domains of helicases were structurally conserved, as well as the protease region and various motifs (6). The seven characteristic motifs of the SPONV helicase appeared in domains 1 and 2, exhibiting a connection with NTP binding and hydrolysis (Fig. 5) (1,2,6,7). One of the most vital motifs in Flaviviridae helicases is the GxGKT/S Motif 1 in domain 1, which is conserved to the same loop in kinases. It is also known as a Walker A motif, and it plays a crucial role in the binding of β-phosphate of ATP (28,29). Based on previous research, the mutagenesis within that motif suggests that the mutant helicase is inactive (29). Furthermore, another important motif for the helicase is the DExH Motif 2, in domain 1. The DExH Motif 2 is responsible for the binding of the Mg2+-ATP substrate, establishing the optimum orientation of ATP for nucleophilic attack (30-32). Finally, although the crucial QRxGRxGR Motif 6 (in domain 2) is not associated with ATP hydrolysis, and its function is exceptionally crucial to the Flaviviridae helicase, as it is involved in nucleic acid binding (7) (Fig. 5).

Figure 4 NS3 Model of the Spondweni virus helicase/protease (numbers 1, 2 and 3 indicate the corresponding domains of the viral helicases). (A) Ribbon representation of the produced Spondweni virus helicase/protease model. (B) Ribbon representation of the produced Spondweni virus helicase/protease model (colored pink) next to the corresponding Dengue virus-4 helicase/protease (colored orange).

Figure 5 The conserved motifs of the Spondweni virus helicase/protease model. The major motifs are color-coded according to the conventions of Fig. 3. (A) Front; (B) back.

The SPONV helicase/protease model, same as all the Flaviviridae helicases, consists of three helicase domains, which are separated by two channels and placed at the C-terminal part of the protein (2). The domains 1 and 2 interact together and to a lesser extent with domain 3. Domain 2 undergoes significant movements compared to the other two domains, during the process of unwinding of double-stranded nucleic acids. The channel between domain 3 and 1-2 accommodates ssRNA during the viral unwinding. RNA binds to the helicase at the arginine-rich site of the 2nd domain (33). Moreover, significant interactions have been identified in the conserved motifs between domains 1 and 2. Motif 3 is necessary in order to stabilize domains 1 and 2. At the same direction, motif 7 forms critical contacts with motifs 1, 2, and 3(2). Last but not least, the protease domain is located in the N-terminal of the model (Figs. 4 and 5).

The model of SPONV was structurally superposed and subsequently compared to its template and with the HCV crystal structure (PDB: 1A1V) (Fig. 6). The alpha-carbon overall RMSD value between the two structures is 0.44 angstroms (Fig. 7). This result confirms the similar configuration of the structures and the good quality of the sequence alignment. However, the RMSD from the superposition of the SPONV model and HCV crystal structure (1A1V) was 3.78 angstroms (Fig. 7). Although RMSD values in the range of RMSD ≤2 indicate unrelated structures, in this case, the value is out of the optimal range due to the lack of the protease domain in the 1A1V structure (Fig. 6). The fact automatically makes the structures non-identical for this specific part of the protein. Moreover, the three viruses, SPONV, DENV, HCV, share similar helicase topology based on the secondary structure features, and a very common active site, as all they belong to the Flaviviridae family. The evaluation of the NS3 SPONV model was performed using the Ramachandran method and the extracted result is provided (Fig. 8) (34). Three amino acids were identified in disallowed regions, due to steric hindrance. A structural analysis was performed in the MOE package in order to examine those amino acids, and the results revealed that they correspond to non-steric hindrance, alpha-helical and beta-sheet conformations. The NS3 SPONV model was also examined for its molecular and electrostatic potential surface (Fig. 9) (35). In the outcome, the results represent the prediction of electrostatically preferred locations of hydrophobic (colored white), H-bond acceptor (colored red) and H-bond donor locations (colored blue). The graphical representation of the electrostatic map provides additional information of the proteins, including potential protein interactions (Fig. 9).

Figure 6 Superposition of the Spondweni virus model (pink ribbon), template structure 2VBC (orange ribbon) and HCV structure 1A1V (blue ribbon). (A) Front; (B) back.

Figure 7 The RMSD plot between the SPONV model, Dengue virus-4 template structure (PDB entry: 2VBC) and hepatitis C virus structure (PDB entry: 1A1V). SPONV, Spondweni virus.

Figure 8 The Ramachandran plot for the Spondweni virus model.

Figure 9 Electrostatic potential surface of the Spondweni virus helicase/protease model and the Dengue virus-4 helicase template (PDB entry: 2VBC) in superposition. (A) Front; (B) back.

Key residues for the design of candidate antiviral agents

The HCV helicase structure 1A1V was used as a starting point in order to identify candidate key residues for the inhibition of the NS3 SPONV helicase. Therefore, the ssDNA molecule was transferred from the HCV structure into the produced NS3 SPONV model, using structural superposition techniques as the MOE. Based on findings from several studies, the channel between domain 3 and 1-2 is a possible candidate inhibition place, since it accommodates the ssRNA during the viral unwinding and consequently plays a critical role in the process of viral replication (3,6,7).

Conserved important residues which are necessary for the helicase activity have been previously identified and have been used as candidate targets for the design of antiviral agents (36,37). Most of these are identified in three important regions of the viral helicase, including Cys431, Arg393 and Arg481. The first region with critical amino acids is the ATP, a pocket-like hydrolysis active site. This region contains an aspartic acid residue, which coordinates the hydrolysis of ATP through its interaction with Mg2+ atoms (3,37). Another important region which may contains key amino acid residues is defined as the core of the domain 3 of Flaviviridae helicases, where the anti-parallel β-sheets impart stability and rigidity, key properties for the functional site of the enzyme. Critical amino acid residues may also exist in the domain, between domain 1 and 2 of the helicase, which constitutes the entrance site of the incoming ssDNA, during the unwinding process (3,36). Therefore, there is increasing interest for the design of candidate inhibitors that will strongly interact with those key residues and regions to block the ssRNA from the entrance into the helicase channel.

Maintaining this strategy, the critical helicase conserved ‘key’ residues were identified in the model as potential targets for structure-based drug design. Using both previous alignment and structural superposition, the residues of Cys429 and Arg388 were identified at the corresponding positions (Figs. 3 and 10). It is obvious that Cys429 residue is a key candidate pharmacological target, being directly exposed to the solvent and strategically located at the center of the viral helicase ssRNA channel. The Cysteine residue should create an S-S or an S-C bond with the inhibitor molecule and to interact with arginine residue, while the Arginine residue is expected to establish H-bond with the inhibitor. The purpose is to develop a bridge between the two conserved residues, Cys429 and Arg388, and also to create a compound that strongly interacts with them, blocking the passage of the ssRNA and consequently the helicase's operation.

Figure 10 The highlighted Cys429 and Arg388 residues of the Spondweni virus helicase model.

In conclusion, computational biology methods have successfully bridged the gap between the lack of experimentally determined protein structures and the design of antivirals, allowing the prediction of a protein's three-dimensional conformation in silico. The three-dimensional structure of the SPONV helicase/protease model was designed using homology modelling techniques. The template structure was the homologous X-ray crystal structure of the DENV-4 helicase/protease (PDB entry: 2VBC), of the same Flaviviridae family. The evaluation of the generated model was successful, exhibiting topological identity to its template. The extensive study of the SPONV helicase/protease model and the identification of key residues provide insight for further studies. In silico methodologies have proven to be an integral part of helicase inhibitor design. The latest Zika outbreak and the ongoing COVID-19 pandemic have highlighted the impact that viral infections can have on global communities and health systems. Therefore, a proactive stance is imperative and the modeling of structures of viral protein targets, such as the Spondweni helicase through computational methods can enable the design of potent antivirals.

Acknowledgements

Not applicable.

Funding

Funding: The authors would like to acknowledge funding from the following organizations: i) AdjustEBOVGP-Dx (RIA2018EF-2081): Biochemical Adjustments of native EBOV Glycoprotein in Patient Sample to Unmask target Epitopes for Rapid Diagnostic Testing. A European and Developing Countries Clinical Trials Partnership (EDCTP2) under the Horizon 2020 ‘Research and Innovation Actions’ DESCA; and ii) ‘MilkSafe: A novel pipeline to enrich formula milk using omics technologies’, a research co-financed by the European Regional Development Fund of the European Union and Greek national funds through the Operational Program Competitiveness, Entrepreneurship and Innovation, under the call RESEARCH-CREATE-INNOVATE (project code: T2EDK-02222).

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

All authors (LP, ET, EP, KID, KP, KD, DAS, FB, GPC, EE and DV) contributed to the conceptualization and design of the study, as well as in the writing, drafting, revising, editing and reviewing of the manuscript. All authors confirm the authenticity of all the raw data. All authors have read and approved the final manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

DAS is the Managing Editor of the journal, but had no personal involvement in the reviewing process, or any influence in terms of adjudicating on the final decision, for this article. GPC is an Editorial Advisor of the journal, but had no personal involvement in the reviewing process, or any influence in terms of adjudicating on the final decision, for this article. The other authors declare that they have no competing interests.

References