Anti-HSV activity of nectin-1-derived peptides targeting HSV gD: an in-silico and in-vitro approach

Abstract

Herpes simplex virus (HSV) infections affect a wide range of the global population. The emergence of resistance to the existing anti-HSV therapy highlights the necessity for an innovative strategy. The interaction of HSV gD with its main host receptor nectin-1 is a potential target for new antiviral drugs. The aim of this study was to develop a peptide derived from nectin-1 targeting HSV gD using the in-silico method and evaluate them for anti-HSV activity. Residues 59-133 of the Nectin-1 V-domain constitute the interaction interface with HSV gD. Bioinformatic tools viz., PEP-FOLD3, ClusPro 2.0, HawkDock and Desmond were used to model the peptide and confirm its binding specificity with HSV gD protein. The peptides with potential interactions were custom synthesized and anti-HSV activity was evaluated in vitro against HSV-1 and HSV-2 by CPE inhibition assay. Five peptide sequences were identified as exhibiting good interaction with HSV-gD proteins. Among them, peptide N1 (residues 76–90) offered maximum protection against HSV-1 (66.57%) and HSV-2 (71.12%) infections. Modification of the identified peptide through peptidomimetic approaches may further enhance the activity and stability of the identified peptide.

Communicated by Ramaswamy H. Sarma

Keywords:

• Anti-HSV peptide
• nectin-1
• HSV gD
• cytopathic effect
• antiviral activity
• antiviral peptide

1. Introduction

Herpes simplex virus (HSV) is an infectious human pathogen that causes life-long latent infections (Baringer & Swoveland, 1973; Singh & Tscharke, 2020). HSV is classified into two distinct types, namely, HSV-1 and HSV-2. HSV-1 is primarily disseminated through oral contact and causes oral herpes or cold sores (Leung & Barankin, 2017). It also causes genital herpes. HSV-2 causes genital herpes via sexual contact. Based on the recently available estimates in 2016, HSV-1 infected over 3.7 billion individuals under the age of 50, which accounts for 66.6% of the worldwide population. HSV-1 infection rates are higher among African (96%) and Middle Eastern and North African (MENA) (89%) populations compared to Latin American and Caribbean (LAC) (85%), Australian (80%), Asian (77%), European (74%), American (58%) and Canadian (51%) populations (Almukdad et al., 2023). HSV-2 affects an estimated 491.5 million (13.2%) people aged 15–49 around the world (James et al., 2020; Looker, Johnston, et al., 2020; Looker, Welton, et al., 2020). Africa has the highest seroprevalence rate of HSV-2 (37%) compared to other regions such as LAC (21%), America (15%), Asia (12%), Europe (12%) and MENA (5%) (Harfouche et al., 2023). It is noteworthy that approximately 187 million (5%) of the global population experiences genital ulcer disease, which is mostly caused by HSV-2 (Looker, Johnston, et al., 2020; Looker, Welton,et al., 2020).

Primary HSV infection begins with the infection of mucosal or skin epithelial cells. The ruptured keratin layer facilitates HSV transmission in the skin (Zhu & Viejo-Borbolla, 2021). HSV replicates in epithelial cells and forms new infectious virions. These virions travel to nerve endings of peripheral neurons (Bello-Morales et al., 2020; Diefenbach et al., 2008; Wilcox & Smith, 1998). By retrograde transport, it reaches the neuronal cell body and establishes latency. Subsequent reactivation of the latent virus results in the production of new virions. These virions undergo anterograde transport toward the skin, leading to herpes lesions (Wilson & Mohr, 2012). Antiviral drugs targeting the replication of HSV, such as Acyclovir (ACV), Famciclovir, Penciclovir, Foscarnet, Cidofovir and Valacyclovir, are currently used to treat HSV infections (Sadowski et al., 2021). Long-term use of these drugs develops resistance in immunocompromised (Epstein & Scully, 1991; Piret & Boivin, 2011) or HIV-infected patients (Ziyaeyan et al., 2007). Vaccines have not been successful in preventing HSV infection so far. The existence of such problems is a growing concern for public health. Thus, the development of new therapeutic approaches that target other mechanisms, such as viral attachment and entry, is needed.

Herpes simplex virus (HSV) has 16 distinct types of surface proteins, of which 12 are classified as glycoproteins. Among these glycoproteins, namely gD, gB, gH and gL are essential in enabling the entry of HSV (Hilterbrand et al., 2021; Huang et al., 2022; Madavaraju et al., 2020). The glycoproteins gB and gH/gL exhibit conservation within the Herpesviridae family (Arii & Kawaguchi, 2018). The gB null mutants demonstrated impaired syncytial cell formation and were incapable of executing cell entry (Dix, 1990). Similarly, deleting either gH, gL or both results in the loss of cell entry and virus-induced cell-to-cell fusion (Forrester et al., 1992; Desai et al., 1988; Hutchinson et al., 1992; Roop et al., 1993). Moreover, glycoproteins gC and gD are conserved in the Alphaherpesvirinae subfamily (Arii & Kawaguchi, 2018; Vega-Rodriguez et al., 2021). The absence of gC (Laquerre et al., 1998; Manservigi et al., 1977) or gD (Fan et al., 2014, 2017; Madavaraju et al., 2020) decreases the entry of free viruses, potentially by reducing viral attachment or binding. HSV gD is reported to play a critical role in HSV attachment and entry (Zhang et al., 2011). The interaction between HSV gD and its cognate receptors (nectin-1, HVEM or 3-OS HS) causes conformational changes in the gD structure and triggers the activation of a fusion protein complex (gB and gH/gL complex), which initiates the fusion process (Fan et al., 2023). Nectin-1 is considered to be the main receptor of gD due to its expression at the productive site of infection (Bhargava et al., 2016; Cocchi et al., 1998; Haarr et al., 2001; Kopp et al., 2009; Petermann et al., 2015; Samanta & Almo, 2015; Wirtz et al., 2020). However, the role of HVEM in the later stages of HSV-1 infection has been postulated to be more important than its involvement in the entry process (Wang et al., 2022). Therefore, it would be significant to target the nectin-1/HSV gD interaction than the HVEM/HSV gD interaction. It is essential to understand the binding interface between nectin-1 and HSV gD in order to effectively target nectin-1/HSV gD interaction. The V-domain of nectin-1 has been found to interact with the HSV gD protein (Cocchi et al., 1998; Giovine et al., 2011; Krummenacher et al., 1999; Menotti et al., 2002; Zhang et al., 2011). Thus, the V-domain of nectin-1 is a promising target for inhibiting the interaction between nectin-1 and HSV gD.

Protein biotherapeutics are a promising approach with the advantages of high target specificity and less toxicity compared to small-molecule drugs. Endogenous proteins that are used as therapeutics are associated with a reduced likelihood of immune responses (Leader et al., 2008). The journey of the Enfuvirtide peptide as an anti-HIV agent highlights the scope of protein- or peptide-based therapeutics for viral infections (Duffalo & James, 2003). Previously, many efforts have been made to treat HSV infection with gC, gB and gH glycoprotein-derived peptides (Cetina-Corona et al., 2016; Franci et al., 2017; Galdiero et al., 2006, 2008; Gianni et al., 2006; Lombardi et al., 2020; Rahangdale et al., 2023; Trybala et al., 2004). Nevertheless, complete protection against infection remains elusive. The 12-mer peptide BP-2 (GSCDGFRVCYMH), derived from two phage-displayed combinatorial peptide libraries, effectively inhibited the binding of HSV-1 gD to HVEM, thereby preventing viral entry into CHO-HVEM cells (Sarrias et al., 1999). However, no peptide derived from the host cell receptor nectin-1 has been reported to block HSV entry, indicating a potential research gap.

Thus, in this study, we have explored peptide therapeutics derived from nectin-1 as a decoy with the potential to inhibit the nectin-1/HSV gD interactions, crucial for HSV entry. Nectin-1-derived peptides were designed and screened for their ability to inhibit HSV infection by in-silico and in-vitro methods.

2. Materials and methods

2.1. In-silico design of novel anti-HSV peptide

2.1.1. Selection of amino acid sequences

The FASTA sequence of the nectin-1 protein was retrieved from the UniProt server (https://www.uniprot.org/) (Primary accession ID: Q15223). Peptide series consisting of 10–20 amino acids were made and then subjected to in-silico analysis for various parameters like charge, hydrophobicity, antigenicity, etc.

2.1.2. Physicochemical properties of peptides

Understanding the physicochemical properties of proteins or peptides enables us to gain insight into their stability, activity and nature. The physicochemical properties of the peptides, such as molecular weight, pI, net charge and hydrophobicity, were predicted by using the ProtParam algorithm of the ExPASy (https://web.expasy.org/protparam/) webserver. PepCalc.com (https://pepcalc.com/) and ToxinPred (http://crdd.osdd.net/raghava/toxinpred/) were used to predict the solubility and toxin profile of peptides, respectively (Gupta et al., 2013).

2.1.3. Peptide structure prediction

The PEP-FOLD3 webserver was used to predict peptide structure from selected amino acid sequences. This web server is based on the hidden Markov model suboptimal sampling approach (Lamiable et al., 2016). The best model with the lowest sOPEP energy (i.e. highest tm value) was selected for further investigation. The predicted structures were validated using the Ramachandran Plot in the PROCHECK server (Laskowski et al., 1993). The protein preparation wizard in the Schrodinger program was used to refine the peptide structure (Schrodinger LLC, New York, NY).

2.1.4. Preparation of target viral proteins

The crystal structures of the target viral proteins HSV-1 gD and HSV-2 gD were retrieved from the Protein Data Bank (HSV-1 gD, PDB ID: 2C36; HSV-2 gD, PDB ID: 4MYV). Protein structures were prepared using the protein preparation wizard in the Schrodinger software suite (Schrödinger LLC). The heteroatoms were deleted, and the structures were altered by assigning the right bond orders, removing water molecules and introducing missing hydrogen atoms. The ionization and tautomeric states of amino acid residues were corrected by adding hydrogen atoms. The structures were minimized to 0.30 Å root mean square deviation (RMSD) using the OPLS3e force field.

2.1.5. Molecular docking and its analysis

Molecular docking of identified peptides with HSV-1 gD and HSV-2 gD was carried out using the ClusPro server (https://cluspro.bu.edu/login.php). Selected protein crystal structures of the peptides as ligands and HSV gD as the receptor were submitted to the ClusPro server (Kozakov et al., 2017). The attraction and repulsion tasks were selected under the ‘Advanced option’ to specify constraints. HSV-1 gD amino residues P23, L25, Q27, R36-H39, Q132, V214-R222, F223, T230-V231 and Y234 were specified in the attraction section of the receptor during peptide/HSV-1 gD docking. Similarly, HSV-2 gD amino residues P23, L25-Q27, R36-H39, Q132, R134, P198, V214-R222, F223, N227, V231 and Y234 were specified in the attraction section of the receptor during peptide/HSV-2 gD docking.

The DIMPLOT server was employed for detailed protein–protein interaction analysis. The Dimplot program schematically illustrates the interaction interface across two protein molecules (Bouzari & Savar, 2014; Forouharmehr et al., 2022). Default criteria were set to determine hydrogen bonds and hydrophobic interactions between the peptide/HSV gD complex. All poses obtained from ClusPro docking were analyzed via DIMPLOT, and the best docking pose was selected based on its interaction with the critical binding interface. The Generalized Born (GB) model and Solvent Accessibility method (MM/GBSA) suite of the HawkDock server was used to rescore poses selected from docking by predicting the binding free energy. The docked complex with the highest negative binding free energy was selected for MD simulation studies (Weng et al., 2019).

2.1.6. Molecular dynamic simulation

The stability of the peptide-protein interactions at HSV gD patches was determined using Desmond software (Schrödinger). The best peptide/HSV gD complex was selected for the MD simulation study. The TIP3P water solvent system with an orthorhombic water boundary box was used. The solvated system was neutralized by adding sufficient sodium or chloride ions and neutralized using 0.15 M NaCl. The peptide/HSV gD complex system was further minimized by the OPLS3e force field at 1 atmospheric pressure and 300 K temperature under the NPT [constant particle number (N), pressure (P) and temperature (T)] ensemble class. Finally, the MD simulation was run for 50 ns. The MD simulation results were analyzed using the simulation interaction diagram tool in the Desmond. RMSD and protein–ligand contact graphs were generated to assess the peptide stability at the HSV gD surface (Thangavel et al., 2020).

2.1.7. MM/GBSA binding free energy calculations

The HawkDock server was used to predict the binding free energy of the peptide/HSV gD complex obtained after MD simulation. The energy-minimized structures of the complex resulting from the MD simulation were uploaded to the MMGBSA suite of the HawkDock server (http://cadd.zju.edu.cn/hawkdock/). Parameters, such as Van der Waals potentials (VDW), electrostatic potentials (ELE), polar solvation-free energies predicted by the GB model, nonpolar contribution to the solvation-free energy calculated by an empirical model (SA) and final estimated binding free energy (kcal/mol) (Total) were calculated.

2.2. In-vitro anti-HSV screening of the designed peptide

2.2.1. Reagents, cells and viruses

Peptide N1 (15-mer peptide: residues 76-90, QNVAIYNPSMGVSVL), and peptide N2 (10-mer peptide: residues 80–89, IYNPSMGVSV) (with purity > 90%) were synthesized by Fmoc based solid phase peptide synthesis method at CDRI Lucknow, India. Peptide N3 (20-mer peptide: residues 61–80, KITQVTWQKSTNGSKQNVAI) (with purity > 95%) (Lot: U882XGJ120-1PE1041), peptide N4 peptide N4 (15-mer peptide: residues 95–109, ERVEFLRPSFTDGTI) (Lot: U213YHF300-5/PE7294) and peptide N5 (16-mer peptide: residues 115–130, ELEDEGVYICEFATFP) (Lot: U586WHJ250-3/PE2936) were custom synthesized by GenScript Inc. (Piscataway, NJ). ACV was purchased from Sigma Aldrich (St. Louis, MO). Thiazolyl blue tetrazolium bromide (MTT) and phosphate-buffered saline (PBS) were procured from HiMedia Laboratories Pvt. Ltd., Mumbai, India.

African green monkey kidney cells (Vero cells) were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) (Himedia) supplemented with 10% fetal bovine serum (Gibco, Grand Island, NY) and 1% antibiotic antimycotic solution (10,000 U penicillin, 10 mg streptomycin and 25 µg amphotericin B per mL) (Himedia) at 37 °C in a 5% CO₂ incubator. HSV 1 (Sc-16) and 2 (133953-03) viruses are being maintained in our laboratory which was procured from Christian Medical College, Vellore, Tamil Nadu.

2.2.2. Solubilization and dilution of peptides

Peptides were dissolved in a suitable solvent based on their nature. Peptide N3, peptide N4 and peptide N5 were weighed and dissolved in sterile distilled water. Peptide N1 and peptide N2 were weighed and dissolved in dimethyl sulfoxide. A stock solution of the peptide was diluted with DMEM without serum. Different concentrations of peptides were prepared with DMEM supplemented with 2% FBS (maintenance medium) for cytotoxicity and antiviral assays.

2.2.3. Cytotoxicity assay

The cytotoxic effect of the synthesized peptides on Vero cells was evaluated by MTT assay. Vero cells were seeded into a 96-well plate at 1 × 10⁴ cells/well density and kept in a CO₂ incubator for 24 h. Hundred µl of different concentrations of each peptide (250, 125, 62.5, 31.25 and 15.62 µg/mL) in maintenance medium were added to quadruplicate wells. The plates were incubated for 72 h at 37 °C in a 5% CO₂ atmosphere and microscopically observed every 24 h of the interval. After 72 h MTT solution (2 mg/mL) and DMSO were added to the cell culture, which stain viable cells and solubilize the formed formazan, respectively. The absorbance of the plate was measured at 540 nm. Data were used to determine % cytotoxicity [% growth inhibition: 100 − (mean OD of individual peptide-treated group/mean OD of control group × 100)] and CC₅₀ (drug concentration that causes 50% cell death) was calculated from the dose-response curve (Denizot & Lang, 1986).

2.2.4. Cytopathic effect (CPE) inhibition assay

The antiviral activity of the custom synthesized peptides was screened for anti-HSV activity by cytopathic effect (CPE) inhibition assay. Vero cells were seeded into a 96-well plate at 1 × 10⁴ cells/well density and kept in a CO₂ incubator for 24 h. Growth media was removed from the microtiter plate upon complete cell confluency and washed twice with PBS solution. This ensures the removal of serum traces from plates prior to HSV exposure. One hundred microliters of HSV-1 or HSV-2 (10 TCID₅₀ dose) virus suspension diluted with maintenance medium was added to each well except for the cell control groups. The plate was incubated for 2 h to allow virus adsorption. Then, the viral suspension from each well was aspirated out, and 100 µL of different non-toxic concentrations of the peptide were prepared in maintenance medium (15.62–250 µg/mL) and added to quadruplicate wells. The standard anti-HSV drug ACV (25 µg/mL) was used as a positive control. One hundred microliters of virus suspension and maintenance medium were added to the virus control and cell control groups, respectively. Then, the plate was incubated for 5 d at 37 °C in a 5% CO₂ atmosphere. Each well of the plate was microscopically observed every 24 h to observe and record the HSV-induced CPE. MTT assay was performed to estimate the percentage of viable cell population after 5 d of incubation, and the resulting data were expressed as a % protection, calculated as [(A – B)/(C – B)] × 100, where A is the absorbance of the peptide-treated infected cells, B is the absorbance of the virus control and C is the absorbance of the cell control (Eisenberg et al., 1980; Lyu et al., 2005).

3. Results

3.1. In-silico design of novel anti-HSV peptide

3.1.1. Selection of amino acid sequences

HSV gD/nectin-1 interaction is crucial for HSV entry (Madavaraju et al., 2020). The HSV gD V domain (residues 56–184) interacts with the Ig-like V domain (residues 31–141) of nectin-1 (Martinez & Spear, 2002; Zhang et al., 2011). Nectin-1 V domain-mapped monoclonal antibodies were able to inhibit HSV gD binding (Krummenacher et al., 2002, 2003). These findings underline the role of the V-domain of nectin-1 in the process of HSV entry. Thus, we selected the nectin-1 V domain to derive peptides for the inhibition of HSV gD binding to the nectin-1 receptor. Nectin-1 amino acid residues S59, K61, T63, Q64, T66, Q68, G73-N77, I80-N82, M85-S88, L90, E125, A127-T131 and N133 of V domain were reported to be involved in HSV-1 gD bindings (Zhang et al., 2011). Therefore, a total of five peptides, i.e. peptide N1 (residues 76–90), peptide N2 (residues 80–89), peptide N3 (residues 61–80), peptide N4 (residues 95–109) and peptide N5 (residues 115–130) were chosen within 59–133 amino acid residues of nectin-1.

HSV-1 and HSV-2 gD have high sequence and structural similarities (∼85%) and interact with nectin-1 at similar domains. (Eisenberg et al., 1980; Lu et al., 2014). HSV-1 gD comprises two nectin-1 binding surface patches, i.e. patch 1 (residues P23, L25, Q27, F223, T230, V231 and Y234) and patch 2 (residues R36, V37, Y38, H39, Q132, V214, D215, S216, I217, G218, M219, L220, P221 and R222). Similarly, HSV-2 gD comprises two surface patches, i.e. patch 1 (residues P23, L25, D26, Q27, F223, N227, V231 and Y234) and patch 2 (residues R36, V37, Y38, H39, Q132, R134, P198, V214, D215, S216, I217, G218, M219, L220, P221 and R222) (Lu et al., 2014; Zhang et al., 2011). The similarity between nectin-1 binding patches of HSV-1 and HSV-2 gD offer a new avenue to develop therapeutics targeting HSV-1 and HSV-2 entry simultaneously. Therefore, peptides derived from nectin-1 were designed with the purpose of binding to the gD glycoprotein of both HSV-1 and HSV-2.

3.1.2. 3D Structure and physicochemical properties of selected peptides

The identified peptides were analyzed for their physicochemical properties using the Prot-Param tool, PepCalc.com, and ToxinPred server. The grand average of the hydropathicity index (GRAVY) is used to predict the hydrophobicity of a peptide. Peptides exhibiting a GRAVY value > 0 are classified as hydrophobic, whereas those with a GRAVY value < 0 are deemed hydrophilic. The peptide N1, N2 and N5 showed 0.613, 0.640 and 0.056 positive GRAVY values, respectively. This indicates that these peptides are highly hydrophobic in nature. On the other hand, the peptides N3 and N4 exhibited a negative GRAVY value, indicating the hydrophilic nature of these peptides. The ToxinPred server predicted that all peptides are non-toxic (Table 1). The 3D structures of the identified peptides and the sOPEP energy of their secondary structures are depicted in Figure 1 and Table 1, respectively.

Figure 1. PEP FOLD-3 predicted 3D structures of nectin-1 derived peptides targeting HSV-gD protein. Peptide structures are presented in cartoon styles with a rainbow color scheme. The color changes from the Nter (dark blue) up to the Cter residue (red-orange).

Table 1. Physicochemical properties of the identified peptides.

Peptide	Sequence	MW (g/mol)	pI	Net charge at pH 7	GRAVY	Solubility in water	Secondary structure information (sOPEP energy)	Toxin prediction
N1	QNVAIYNPSMGVSVL	1591.84	5.52	0	0.613	Poor	−18.4033	Non-toxin
N2	IYNPSMGVSV	1066.24	5.52	0	0.640	Poor	−9.30933	Non-toxin
N3	KITQVTWQKSTNGSKQNVAI	2231.54	10.30	+3	−0.750	Good	−17.4852	Non-toxin
N4	ERVEFLRPSFTDGTI	1766.97	4.68	−1	−0.373	Good	−12.6639	Non-toxin
N5	ELEDEGVYICEFATFP	1862.04	3.45	−5	0.056	Good	−12.9467	Non-toxin

Molecular weight (MW), Isoelectric point (pI), net charge and Grand average of hydropathicity index (GRAVY) were calculated by using the ProtParam tool; Solubility was predicted by PepCalc.com-peptide calculator; Optimized Potential for Efficient structure Prediction score (sOPEP) of peptide structure was obtained from the PEP FOLD3 webserver.

Table 2. Balance weighted score of the best pose of peptide/HSV gD complex.

Peptide/HSV gD complex	Balance weighted score
	Center energy	Lowest energy
N1/HSV-1 gD	−852.0	−923.1
N1/HSV-2 gD	−796.3	−831.6
N2/HSV-1 gD	−636.9	−636.9
N2/HSV-2 gD	−560.3	−576.6
N3/HSV-1 gD	−579.4	−589.5
N3/HSV-2 gD	−777.9	−901.7
N4/HSV-1 gD	−880.2	−918.9
N4/HSV-2 gD	−837.6	−901.3
N5/HSV-1 gD	−861.7	−899.6
N5/HSV-2 gD	−841.8	−854.2

The balanced weighted score represents the PIPER energy scoring function generated by ClusPro version 2.0. The ‘center energy’ and ‘lowest energy’ are used to describe the energy of the central structure and the structure with the lowest energy within a cluster, respectively. Both metrics indicate the quality of predicted protein complex structures and do not specify binding affinity.

3.1.3. Molecular docking and its analysis

The docking study aimed to identify the peptide interactions at nectin-1 binding HSV gD patches. The ClusPro docking server predicted 5–10 poses for each peptide/HSV gD complex, and these were primarily narrowed down using comprehensive DIMPLOT analysis. The binding free energies of the top poses were calculated. The model scores obtained after docking of the designed peptides with HSV-1 gD and HSV-2 gD proteins are shown in Table 2.

Docking results showed that all five designed peptides bind to nectin-1 interacting patches of HSV-1 gD and HSV-2 gD proteins. Peptide N1 residues Tyr6 and Asn7 were found to interact with the Ile217 and Arg222 residues of HSV-1 gD patch 2. Moreover, peptide N1 also bound to Asp215, Ile217 and Tyr38 residues of HSV-2 gD patch 2 through residues Ala4, Gln1 and Asn7, respectively (Figure 2). Peptide N2 residue Asn3 interacts with Arg222 residue of HSV-1 gD patch 2 region. It also interacts with the Asp215, Val37 and Arg134 residues of HSV-2 gD patch 2 through residues Ile1, Tyr2 and Asn3, respectively (Figure S1). Gln16, Asn17, Lys1 and Lys15 residues of peptide N3 bound to Arg222 and Ile217 residues of the HSV-1 gD patch 2. Whereas peptide N3 residues Asn17 and Ser14, Asn12, Val18 and Ala19, Ala19, Ile20 and Lys9, interact with the Gln132, Arg222 and Pro221, Arg134, Tyr38, Val37 and Asp215 residues of HSV-2 gD patch 2, respectively (Figure S2). Peptide N4 residue Arg2 showed interaction with the Thr230 residue of HSV-1 gD patch 1. Peptide N4 residues Asp12 and Gly13 interact with the His39 residue of HSV-2 gD patch 1. It also interacts with the Arg134, Val214 and Asp215 residues of HSV-2 gD patch 2 through Gly13, Thr14, Ile15 and Thr11 residues (Figure S3). Another peptide N5 bound to Arg222 residues of HSV-1 gD patch 1 through Tyr8 and Glu11 residues. Glu1 and Glu11 residues of this peptide interacted with the Ser216 and, His39 and Arg134 residues of HSV-2 gD patch 2, respectively (Figure S4). This demonstrated that the peptides derived from nectin-1 have the ability to bind to HSV gD and can disrupt the interaction between nectin-1 and HSV gD.

Figure 2. Interaction of peptide N1 with (a) HSV-1 gD protein and (b) HSV-2 gD protein. The protein/peptide complex in surface representation illustrates contacts between HSV gD protein (in blue color) and peptide (in magenta color). The dotted red color box shows a close-up of the docking area (LigPlot DIMPLOT interaction). Amino acid residues of HSV gD protein and peptide are represented with blue and magenta colors, respectively. Green lines represent hydrogen bonds, and bond lengths are labeled on the lines. The critical interactions of peptide N1/HSV gD complex are highlighted with a boutique beige color.

3.1.4. Molecular dynamic simulation

The stability of the peptide interaction with HSV gD protein was analyzed through MD simulation study. Interpretation of the RMSD plot and ligand-protein contacts (2D interaction diagram) was carried out to gain a comprehensive understanding of the simulation outcomes. The RMSD plot represents the RMSD evolution of the HSV gD protein (left y-axis) and ligand (peptide) (right y-axis). In the RMSD plot, protein frames were aligned on the reference frame backbone and displayed changes in their structural conformation throughout the simulation. Ligand (peptide) RMSD illustrates ligand stability relative to the protein frame. 2D-trajectory interaction diagram showing % simulation time of peptide interaction with the target protein residues. Interactions that take place for >30.0% of the simulation time are represented in a 2D-trajectory interaction diagram. The percent simulation time reflects the strength of the interaction. Hydrogen bonds retaining for > 30%, 40–70% and > 70% of the simulation time are considered weak, medium, and strong interactions, respectively.

The protein and ligand (peptide) RMSD plots of a peptide N1/HSV-1 gD complex and, peptide N1/HSV-2 gD complex (Figure 3) remained stable until the end of the simulation with RMSD values less than 3 Å. In MD simulation, the interaction of peptide N1 residues with Arg222 of HSV-1 gD that was observed during docking remains stable. The critical residue Arg222 interacts with the oxygen of the -NH-C = O group through a hydrogen bond for 99% of the total simulation time. It also formed the cation–π interaction with the Tyr6 residue. The Arg222 residue exhibited a strong interaction and was essential for peptide binding (Figure 4). Another interaction of peptide N1 with Ile217 residues of HSV-1 gD occurred for <30% of the total simulation time and lost the contacts in a dynamic state. Similarly, it was observed that the peptide N1 residues lost contact with the Asp215, Ile217 and Tyr38 residues of the HSV-2 gD patch 2. Additionally, another critical residue Gln132 of HSV-2 gD bound to the oxygen of the -NH-C = O group through two hydrogen bonds for 44% and 47% of the total simulation time (Figure 5). This interaction exhibited a moderate level of strength. Thus, peptide N1 interaction with Arg222 residue of HSV-1 gD patch 2 and Gln132 residues of HSV-2 gD patch 2 was found to be stable in MD simulation.

Figure 3. RMSD plot of the peptide N1 with HSV-1 gD and HSV-2 gD protein.

Figure 4. 2D diagram of ligand‒protein interaction of the peptide N1/HSV-1 gD complex.

Figure 5. 2D diagram of ligand‒protein interaction of the peptide N1/HSV-2 gD complex.

RMSD plot of the peptide N2/HSV-1 gD complex and peptide N2/HSV-2 gD complex (Figure S5) shows slight fluctuation throughout the 50 ns simulation. Ligand deviation was observed in both the peptide N2/HSV-1 gD complex and the peptide N2/HSV-2 gD complex. This suggests ligand conformational changes at the protein-interacting site. However, the RMSD deviation was within an acceptable range of 3 Å. The critical residue Arg222 of HSV-1 gD formed two hydrogen bonds for 62% and 86% of the total simulation time. Thus, this interaction was found to be strong and important for peptide binding (Figure S6). On the other hand, the interaction between peptide N2 and critical residues Asp215 and His39 of HSV-2 gD was observed for 42% and 30%, and 31% of the total simulation time, respectively (Figure S7). Thus, the interaction of peptide N2 with residues Asp215 and His39 of HSV-2 gD was found to be weak.

The RMSD plot of the peptide N3/HSV-1 gD complex was very stable over the entire simulation period (Figure S5). However, critical residues Arg222 (32%) and Ser216 (35%) of HSV-1 gD showed a weak strength of interaction. The strength of this interaction was found to be sufficiently weak to ensure stable contact (Figure S8). On the other hand, it is worth noting that despite the fluctuation in protein and ligand RMSD of peptide N3/HSV-2 gD complex during the simulation, the essential interaction remained consistently stable (Figure S5). Residues Pro221 (73%), Gln132 (63% and 79%) and Arg134 (87% and 46%) of HSV-2 gD exhibited a strong interaction (Figure S9).

The protein and ligand RMSD plot of the peptide N4/HSV-1 gD complex and peptide N4/HSV-2 gD complex (Figure S10) was stable throughout the 50 ns simulation period. Contact between peptide N4 and critical residue Thr230 (49%) of HSV-1 gD demonstrates the moderate strength of interaction (Figure S11). On the other hand, peptide N4 exhibited strong contact with His39 (61%), Arg134 (76% and 62%) and Gln132 (97%) residues of HSV-2 gD (Figure S12). The RMSD plot of a peptide N5/HSV-1 gD complex and peptide N5/HSV-2 gD complex (Figure S10) shows fluctuation; however, the fluctuation was within the 3 Å range. Residue Arg222 (41% and 45%) and Ser216 (34% and 46%) of HSV-1 gD and Arg222 (45% and 47%) of HSV-2 gD demonstrate a moderate level of strength (Figures S13 and S14).

3.1.5. MM/GBSA binding free energy calculations

The binding free energies of the peptide/HSV gD complexes were estimated using MM/GBSA analysis. The MM/GBSA binding free energies of each peptide/HSV gD protein complexes are given in Table 3. Among each peptide/HSV-1 gD complex, peptide N5/HSV-1 gD complex showed the lowest binding free energy value (−105.19 Kcal/mol). In contrast, the peptide N2/HSV-2 gD complex exhibited the lowest binding free energy value (−95.82 Kcal/mol) among other peptide/HSV-2 gD complexes.

Table 3. MM-GBSA binding free energies of each peptide/HSV gD complex.

Peptide − HSV gD complex		VDW	ELE	GB	SA	Total (Kcal/mol)
N1	HSV-1 gD	−132.97	−106.96	160.35	−16.42	−96
	HSV-2 gD	−56.49	−43.49	69.95	−7.57	−37.61
N2	HSV-1 gD	−95.82	−97.57	133.34	−12.26	−72.32
	HSV-2 gD	−133.18	−106.57	160.35	−16.43	−95.82
N3	HSV-1 gD	−145.14	−402.08	475.53	−17.51	−89.2
	HSV-2 gD	−44.08	−232.26	249.29	−5.26	−32.31
N4	HSV-1 gD	−49.34	−141.69	159.64	−7.29	−38.68
	HSV-2 gD	−78.72	19.79	12.9	−9.46	−55.49
N5	HSV-1 gD	−120.81	−7.65	40.22	−16.95	−105.19
	HSV-2 gD	−39.49	−16.62	22.87	−7.3	−40.54

The binding free energy of the complex was calculated by using the HawkDock server.

VDW: Van der Waals potentials; ELE: electrostatic potentials

Polar Solvation free energies predicted by the GB model; Nonpolar contribution to the solvation-free energy calculated by an empirical model (SA); Final estimated binding free energy (kcal/mol) calculated from VDW, ELE, GB and SA (total).

From the docking study, it was observed that all the peptides bind to one of the critical residues of the nectin-1 binding HSV gD patches. Further, MD simulation data revealed that most of the peptides were found to maintain those interactions and encountered new interactions by other critical residues of nectin-1 binding HSV gD patches. Disrupting the nectin-1/HSV gD interaction was a target for these designed peptides. The binding potential of all designed peptides to the HSV gD was demonstrated through docking, molecular dynamics (MD) simulation data, and binding free energy calculations. Subsequently, the validity of these peptides was further confirmed through in-vitro experiments.

3.2. In-vitro screening of the designed peptide

3.2.1. Cytotoxicity studies

The cytotoxic effect of peptides on Vero cells was investigated by MTT assay. All designed peptides were nontoxic to Vero cells at concentrations between 15.62 and 250 µg/mL (Figure 6(a)).

Figure 6. (a) The effects of various concentrations of test peptides on the viability of Vero cells. Cell viability was measured by MTT assay. Percentage viability data are representative of the results of three independent experiments and expressed as the means ± SDs; Antiviral activity of the designed peptides against (b) HSV-1 infection and (c) HSV-2 infection. Percentage protection data are representative of the results of three independent experiments and expressed as the means ± SDs. Data were plotted using Graph Pad Prism software version 8.0.2.

3.2.2. Anti-HSV activity of the designed peptides

A CPE inhibition assay was carried out to evaluate the anti-HSV activity of the designed peptides. CPE was characterized by cell rounding, swelling, cell detachment and ultimately cell death. Five nontoxic concentrations of the peptides ranging between 15.62 and 250 µg/mL were tested against 10 TCID₅₀ virus challenge dose. The standard anti-HSV drug ACV showed 100% protection against both HSV-1 and HSV-2 infections when treated at 25 µg/mL (data not shown). MTT assay was performed to calculate the percentage of viable cell population in the treated cells.

Peptide N1 showed dose depended protection, i.e. 66.57%, 50.6% and 34.55% against HSV-1 infection at 250 µg/mL, 125 µg/mL and 62.5 µg/mL concentrations, respectively (Figure 6(b)). On the contrary the peptides N2, N3, N4 and N5 showed ≤ 40% protection against HSV-1 infection at the highest tested concentration of 250 µg/mL. The antiviral activity of these peptides against HSV-2 infection is shown in Figure 6(c). The peptide N1 showed 71.12% and 56.99% protection against HSV-2 infection at 250 µg/mL and 125 µg/mL concentrations, respectively. Similarly, peptide N2 showed 65.77% and 48.72% protection against HSV-2 infection at 250 µg/mL and 125 µg/mL concentrations, respectively. The peptides N3, N4 and N5 showed ≤ 40% protection against HSV-2 infection at the highest tested concentration of 250 µg/mL.

4. Discussion

Nectin-1 and HSV gD protein interaction is a new avenue to target the HSV entry (Lu et al., 2014; Zhang et al., 2011). This work firmly confirms critical regions on the nectin-1 which are crucial for the HSV gD protein interaction. Nectin-1 is mainly expressed in epithelial, neuronal and endothelial cells and is considered as a primary receptor (Geraghty et al., 1998; Simpson et al., 2005). The binding of HSV gD to nectin-1 initiates HSV entry and causes conformational changes in the gD structure, triggering the fusion process (Lazear et al., 2014). Thus, the nectin-1/HSV gD interaction that governs HSV entry is an ideal target for novel antiviral agents to block HSV entry.

The antiviral peptide was designed based on several criteria that are presumed to be important for the inhibition of the nectin-1/HSV gD interactions. First, the binding interface between nectin-1 and HSV gD interaction was identified. The ectodomain of the HSV gD protein consists of the Ig-like V-domain (residues 56–184) flanked by N-terminal (residues P23 − I55) and C-terminal (residues A185–A253) extensions. These terminal extensions act as a nectin-1 binding site of HSV gD. As mentioned previously, the essential amino residues of HSV gD belonging to these extensions are divided into patch 1 and patch 2 (Zhang et al., 2011). On the other hand, the ectodomain of nectin-1 consists of three Ig-like domains (i.e. one V-set Ig and two C-set Ig domains). It has been reported that nectin-1 exploits its Ig-like V domain to contact HSV gD. The gist of these studies is that the V-domain (residues 56–184) of HSV gD binds with the V-domain (residues 31–141) of nectin-1 (Carfí et al., 2001; Kwon et al., 2006; Martinez & Spear, 2002). Hence, we hypothesized that V-domain-derived peptides would interact with HSV gD and thereby function as competitive inhibitors that can prevent the nectin-1/HSV gD interaction, thus preventing viral cell entry.

Thus, peptides from the V-domain of nectin-1 were identified that can bind to HSV gD. Here, V-domain amino residues S59, K61, T63, Q64, T66, Q68, G73-N77, I80-N82, M85-S88, L90, E125, A127-T131 and N133 of nectin-1 that interact with HSV gD (Zhang et al., 2011) were considered while selecting peptide sequences. Amino acid conservation or similarity between HSV-1 and HSV-2 gD (∼85%) was also considered while designing the peptides so that they can bind to either HSV-1- or HSV-2 (Eisenberg et al., 1980; Lu et al., 2014).

A docking strategy and MD simulation study were applied to evaluate the potential binding of the designed peptides at the desired site on HSV gD. Our findings suggested that all designed peptides interact with one of the critical residues located at patch 1 or patch 2 of HSV gD (Lu et al., 2014; Zhang et al., 2011). Table 4 illustrates a summary of the predicted critical interactions established between peptides and the HSV gD protein. The in-silico findings were validated through an in-vitro antiviral assay. Peptide N1 demonstrates maximum antiviral activity against both HSV-1 and HSV-2 infections at a concentration of 250 µg/mL (equal to 157.05 µM). A higher affinity between two protein partners results in a more stable complex with the lowest binding free energy. The second lowest binding free energy of peptide N1/HSV-1 gD complex (−96 Kcal/mol) indicates the higher affinity of peptide N1 for HSV-1 gD protein. The peptide formed robust interactions with the Arg222 residue of HSV-1 gD via hydrogen bonding and cation–π interactions. Cation–π interaction at the protein–protein interface is favorable to drug binding. Cation–π interactions with Arg significantly contribute to the binding energy of protein complex formation (Crowley & Golovin, 2005; Dougherty, 2013; Zhong et al., 1998). Moreover, cation–π pairs interaction involving arginine also tends to form hydrogen bonds. These interactions are anticipated to ensure the binding specificity (Crowley & Golovin, 2005). Thus, cation–π and hydrogen bond interaction between the peptide residue Tyr6 and the Arg222 residue of HSV-1 might impede the nectin-1/HSV-1 gD interaction, leading to a reduction in HSV-1 infection.

Table 4. Summary of the interactions identified between peptides and HSV gD protein.

Peptide	Interactions identified by docking study		Interactions identified by MD simulation (2D interaction diagram data)
	HSV-1 gD Peptide–protein residue (no. of HB)	HSV-2 gD Peptide–protein residue (no. of HB)	HSV-1 gD (% simulation time)	HSV-2 gD (% simulation time)
N1	Tyr6 – Arg222 (2), Tyr6 – Ile217 (1), Asn7 – Arg222 (1)	Gln1– Ile217 (1), Ala4 – Asp215 (1), Asn7 – Tyr38 (1)	Arg222 (99%, 39% and 32%)	Gln132 (44% and 47%)
N2	Asn3 – Arg222 (2)	Ile1 – Asp215 (1), Tyr2 – Val37 (1), Asn3 – Arg134 (1)	Arg222 (62% and 86%)	His39 (31%), Asp215 (30% and 42%)
N3	Lys1 – Ile217 (1), Lys15 – Ile217 (1), Gln16 – Arg222 (1), Asn17 – Arg222 (1)	Lys9 – Asp215 (1), Asn12 – Pro221 (1), Asn12 – Arg222 (1), Asn12 – Gln132 (1), Ser14 – Gln132 (1), Asn17 – Gln132 (1), Val18 – Arg134 (1), Ala19 – Arg134 (1), Ala19 – Tyr38 (1), Ile20 – Val37 (1)	Arg222 (32%) and Ser216 (35%)	Gln132 (63%), Arg134 (87%) and Pro221 (73%)
N4	Arg2 – Thr230 (1)	Gly13 – Arg134 (1), Ile15 – Arg134 (1), Thr14 – Arg134 (1), Thr14 – His39 (1), Asp12 – His39 (2), Thr11 – Val214 (2), Thr11 – Asp215 (2)	Thr230 (49%)	His39 (61%), Arg134 (76% and 62%), Gln132 (97%)
N5	Tyr8 – Arg222 (1) Glu – Arg222 (1)	Glu1 – Ser216 (1), Glu11 – Arg134 (2), Glu11 – His39 (1, salt bridge)	Ser216 (34% and 46%), Arg222 (41% and 45%)	Arg222 (45% and 47%)

HB: hydrogen bond

On the other hand, peptide N1 demonstrated the second lowest affinity for HSV-2 gD among other peptides and exhibited moderate strength of interaction with the Gln132 residue of HSV-2 gD. Regardless of this factor, peptide N1 demonstrated better protection against HSV-2 infection. It is noteworthy that despite the relatively weak hydrogen bond contacts, such interactions are essential for stabilizing the conformations of the ligand–protein complex (Bissantz et al., 2010). This could be a contributing factor to the notable antiviral activity of the peptide N1 against HSV-2.

Peptide N2 demonstrated moderate interaction with critical residues His39 and Asp215 of HSV-2 gD. However, it exhibited the highest binding affinity for HSV-2 gD compared to other peptides and possessed notable antiviral activity against HSV-2 infection at 250 µg/mL (equal to 234.46 µM). In contrast, peptide N2 failed to exhibit significant activity against HSV-1. Sequence homology of HSV gD type was considered while designing peptides. All the peptides tested in this study showed varying degrees of antiviral activity against HSV-1 and HSV-2. Such variability in antiviral activity has also been noted with peptides derived from HSV gH and gB glycoproteins (Cetina-Corona et al., 2016).

Other peptides, such as peptide N3, peptide N4, and peptide N5 were tested at the highest 250 µg/mL concentration (equal to 112.03, 141.48 and 134.26 µM, respectively). These peptides did not show noticeable antiviral activity against either HSV-1 or HSV-2 infections. Irrespective of the lowest binding free energy of peptide N5/HSV-1 gD complex (−105.19 Kcal/mol) compared to a peptide N1/HSV-1 gD complex, peptide N5 failed to illustrate prominent antiviral activity against HSV-1 infection. Similar trends have been noted in relation to the aptamer that targets HSV gD. Aptamers exhibiting a strong affinity for HSV gD failed to show anti-HSV activity evaluated by in-vitro methods (Gopinath et al., 2012). Our peptide modeling studies showed that these peptides did not show a lopped structure like N1 and N2 (Figure 1). We believe that, in addition to the biophysical parameters, the looped structure is needed to bind to the HSV-1 gD complex more rigidly.

The hydrophobicity of the peptide may be another contributing factor behind the observed variation in antiviral activity. Hydrophobic peptides are believed to possess effective antiviral activity against enveloped viruses (Badani et al., 2014). This hypothesis is supported by the notable antiviral activity of hydrophobic peptides N1 and N2, as well as the comparatively lower antiviral activity of hydrophilic peptides N3 and N4. Peptide N5 with lower antiviral efficiency is less hydrophobic due to a lower % hydrophobic residue content (50%) compared to peptide N1 (60%) and peptide N2 (60%). Thus, these findings depict the correlation between the hydrophobicity of the peptides and their antiviral potency. Similar trends of results were shown by the hydrophobic 12-mer FluPep peptide and 16-mer EB peptide against the influenza virus. These peptides were anticipated to engage in interactions either with their own kind or with membranes of both cellular and viral origin (Jones et al., 2006; Nicol et al., 2012). These hydrophobic peptides exhibit favorable bioavailability as a result of their ability to bind and translocate across membranes (Badani et al., 2014).

In-silico and in-vitro results of the peptide N1 suggested that there might be a direct interaction between nectin-1 V-domain-derived peptide and HSV gD protein causing the inhibition of the nectin1/HSV gD interaction. Consequently, this led to a reduction in both HSV-1 and HSV-2 infections. Additional research is required to validate the precise mechanism of the interaction between the peptide and HSV gD. The surface plasmon resonance method can be used to investigate these interactions at a molecular level (Cairns et al., 2019; Douzi, 2017; Gopinath et al., 2013).

5. Conclusion

In this study, we have demonstrated that the interaction of nectin-1/HSV gD involved in the entry process can be a good target for designing antiviral peptides. This is the first study to report the antiviral activity of nectin-1-derived peptides targeting HSV gD. Two out of five nectin-1-derived peptides selected based on the in-silico results demonstrated good anti-HSV activity in vitro. The peptide N1 showed consistent antiviral activity against both HSV-1 and HSV-2 infections. Though the antiviral activity exhibited by this peptide is lower than the standard drug ACV, it is a step forward toward finding alternative anti-HSV therapeutics. These findings emphasized the antiviral potential of the 76 − 90 region of the V-domain of nectin-1. The therapeutic activity of this peptide can be enhanced through peptidomimetic approaches. This finding could provide a structural framework for further development of anti-HSV therapeutics.

Author contributions

Mr. Rakesh Rahangdale and Ms. Parnavi Ghormode performed the experiments, analyzed the data, and wrote the draft manuscript; Mr. Tenzin Tender, Ms. Sridevi Balireddy, Dr. Fayaz Shaik Mohammad, Mr. Raj Kishore and Mr. Sumit Birangal contributed to analyzing the data and helped in the preparation of the manuscript. Dr. Raghu Chandrashekar H, Dr. Mukesh Pasupuleti and Mr. Rakesh Rahangdale conceptualized the study. Dr. Raghu Chandrashekar H. supervised the study. All the authors reviewed the manuscript and approved the contents of the manuscript.

Abbreviations
HSV	=	Herpes simplex virus
MENA	=	Middle Eastern and North African
LAC	=	Latin American and Caribbean
ACV	=	Acyclovir
RMSD	=	Root mean square deviation
MM/GBSA	=	Molecular mechanics with generalized Born and surface area solvation
VDW	=	Van der Waals potentials
ELE	=	Electrostatic potentials
MT T	=	Thiazolyl blue tetrazolium bromide
P BS	=	Phosphate-buffered saline
CP E	=	Cytopathic effect
GRAVY	=	Grand average of the hydropathicity index
sOP EP	=	Optimized Potential for Efficient Structure Prediction score
MW	=	Molecular weight
pI	=	Isoelectric point
MD	=	Molecular dynamics

Acknowledgments

We are thankful to the Indian Council of Medical Research (ICMR), New Delhi, for providing an ICMR-SRF fellowship to Mr. Rakesh R. Rahangdale (File No-5/3/8/56/ITR-F/2022-ITR), Mr. Tenzin Tender (File no-2020-6096/PROTEOMICS-BMS) and Ms. Sridevi Balireddy (File no-45/14/2020-DDI/BMS). We also gratefully acknowledge Manipal College of Pharmaceutical Sciences, MAHE, Manipal, India, for providing the necessary facilities. We also express gratitude to the Manipal-Schrodinger Centre for Molecular Simulation.

Disclosure statement

The authors report there are no competing interests to declare.

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article and its Supplementary Materials.

Additional information

Funding

This research work was supported by an Intramural fund MAHE/DREG/PHD/IMF/2019 dated 7 February 2019 (UTN: RG0619259) grant.