Background and Aim: Cervical cancer, primarily caused by human papillomavirus (HPV), remains a significant health challenge. High-risk HPV types, particularly through the persistent expression of oncoproteins E6 and E7, disrupt tumor suppressor functions. Recognizing the role of Heat Shock Protein 110 (HSP110) in enhancing immune responses, this study explores the design and therapeutic potential of a novel T-cell-epitope-based DNA vaccine targeting these viral components.

Materials and Methods: We constructed a chimeric HPV16 E6/E7/HSP110 vaccine with high molecular weight, a basic nature, good aqueous solubility, and prolonged stability/half-life. We investigated immune reactivity, conservancy among HPV strains, and affinity to certain HLA types for potential antigenic peptides emanating from E6, E7, and HSP110. To this end, the three-dimensional structure of this protein was further refined using server-based methods and its interaction with HLA class I [HLA-A11:01] and MHC-II molecules [HLA-DRB1_01:01] was assessed by in silico molecular docking studies. Additionally, we performed a cell-based ELISA on HPV16-transformed TC-1 target cells using sera from healthy vaccinated mice to assess antigen-specific binding to cellular targets; cytokines were quantified separately by sandwich ELISA on culture supernatants as a complementary readout.

Results: The T-cell epitopes of HPV16 E6, E7, and HSP110 were predicted, and 10 epitopes with strong antigenic properties and non-toxicity were selected. These epitopes were linked using a KK linker to generate a multi-epitope vaccine. The resulting vaccine exhibited high antigenicity, non-allergenicity, and good solubility with favorable physicochemical properties. The 3D structure of the vaccine and its interactions with HLA class I [HLA-A11:01] and class II [HLA-DRB1_01:01] were successfully demonstrated. Molecular dynamics (MD) simulations confirmed the stability of the vaccine binding to HLA-A11:01. Codon optimization and virtual cloning predicted high expression in E. coli K12; in vitro, Western blotting in HEK293T cells confirmed expression of the chimeric construct. Preliminary cell-ELISA indicated increased secretion of selected cytokines (e.g., IFN-γ, IL-2, TNF-α) relative to appropriate controls, supporting cell based immunogenicity.

Conclusion: The data collected demonstrated that the proposed construct represents a novel and promising therapeutic vaccine candidate for cervical cancer. Although further validation with additional experimental work and clinical trials is needed, our study provides a starting point to guide the investigation of new treatment strategies for cervical cancer.

• Cervical cancer

• Human papillomavirus

• Therapeutic vaccine

• Immunoinformatics

• Antigenic peptides

• Heat shock protein 110

Fazeli M, Mostafazadeh B, Afzadi HA, Kafi ZZ, Zarmehri B, et al. (2026) Expression Analysis and Preliminary Cell-Based Immunogenicity of a Novel T-Cell Epitope-Driven DNA Vaccine Candidate against Cervical Cancer in Eukaryotic Cells. JSM Biol 8(1): 1023.

Cervical cancer is the fourth most common cancer in women globally [1]. Human papillomavirus (HPV) is the principal etiological agent; high-risk genotypes express the oncoproteins E6 and E7 that disable tumor suppressors [1]. Beyond cervical cancer, HPV is implicated in anal, vulvar, vaginal, and penile cancers, as well as a subset of head and neck cancers, particularly oropharyngeal cancers. Chronic infection with high-risk types, mainly HPV16 and HPV18, accounts for ~70% of cervical cancers [2,3].

There are a significant number of people worldwide with high-risk HPV infections and related conditions. The design and development of a therapeutic vaccine are crucial due to several factors: (1) Prophylactic vaccines may take over 20 years to show effects in preventing cervical cancer, as precancerous lesions develop slowly [4]. (2) L1 and L2 proteins, which are the basis of current prophylactic vaccines, are not expressed in infected basal epithelial cells, limiting their therapeutic impact [5]. (3) While most HPV infections clear spontaneously within two years in about 90% of cases [6], persistent high-risk infections necessitate targeted therapeutic interventions [6].

There remains a clinical need for appropriate, effective therapeutic options in patients already infected with high-risk HPV strains. Prophylactic vaccination has reduced the incidence of new infections, but it offers no solution for those individuals already infected or those in whom pre-neoplastic or neoplastic lesions have arisen [7]. Accordingly, we developed a DNA vaccine to bridge this gap by eliciting robust cellular immunity capable of eliminating infected cells. Incorporating HSP110 is expected to enhance antigen presentation and therefore improve the effectiveness of the immune response against HPV-infected cells [8]. This strategy addresses an unmet clinical need not met by prophylactic vaccines.

Multi-epitope vaccines (MEV) are recombinant vaccines considered a promising strategy against tumors and viral infections due to their high specificity, safety, stability, and low-cost production [9]. Although multi epitope vaccines may exhibit limited immunogenicity due to rapid peptide proteolysis in vivo and suboptimal uptake by antigen-presenting cells [10], one strategy to augment the immune response elicited by multi-epitope constructs is to leverage heat shock proteins (HSPs), which can enhance antigen delivery and T-cell responses [10-12]. HSP110 is an established elicitor of innate and adaptive immunity in cancer and infectious-disease vaccines and can function as an adjuvant in their development [13].

Immunoinformatics has the potential to significantly accelerate the vaccine-development process by predicting epitopes capable of binding to major histocompatibility complex (MHC) molecules and stimulating T cells, which is crucial for successful anti-tumor immune responses. Several studies have illustrated that peptides derived from HSPs can induce immune responses; for example, HSP70-based peptides have stimulated T-cell responses in tumor models [13,14]. These observations support our use of HSP110 peptides in a multi-epitope vaccine for cervical-cancer immunotherapy.

Several studies have been conducted to develop a therapeutic HPV vaccine. For example, multi-epitope constructs comprising E6/E7 T-cell epitopes have been evaluated in vivo and shown evidence of inhibiting tumor growth [15]. Another study found that the E6E7-VEGFR2 combination vaccine had a more significant anti-tumor effect than E6E7 alone [16].

Currently, no therapeutic vaccines for HPV are approved for human use, but extensive research and clinical trials are ongoing [17,18]. In this study, we aimed to develop a new multi-epitope vaccine against HPV16 using bioinformatics. We focused on the E6 and E7 oncoproteins, known for their role in cervical carcinogenesis, and HSP110, which enhances immune response. Epitopes were selected based on their low toxicity, non-allergenicity, and high immunogenicity, and were linked using a KK linker to optimize the vaccine’s potential effectiveness. We assessed the vaccine’s binding affinity to HLA class I (HLA-A*11:01) and class II (HLA-DRB1*01:01) molecules using docking simulations. In addition, we modeled and refined the vaccine’s three-dimensional structure. To complement the in-silico analyses, we also conducted preliminary cell-ELISA–based cytokine profiling in non-malignant cells transfected with the construct to provide an initial cell-based immunogenicity readout (Figure 1).

Figure 1: A diagram illustrating the immunoinformatic approaches to developing a vaccine against several epitopes.

In Silico Analysis

Sequence Retrieval: Data Source and Accession Numbers: We obtained the amino acid sequences for HPV16 E6 (NP_041325.1), E7 (NP_041326.1), and human HSP110 (NG_053051.1) from the NCBI database, all in FASTA format. Using immunoinformatic tools, we aimed to predict and select epitopes based on their antigenicity, immunogenicity, conservation, and ability to bind to HLA alleles.

Construct design: Antigenic Peptide Prediction: Using computational tools such as the Immune Epitope Database (IEDB) and NetMHC (NetMHCpan-4.1 and NetMHCIIpan-4.0), potential T-cell epitopes were identified from the HPV16 E6, E7, and HSP110 proteins. IEDB was used to make predictions on the potential epitopes that may be involved in immune responses. This tool helps researchers identify epitopes with a high potential to bind with HLA molecules, thus stimulating the immune response. The epitope binding prediction tool NetMHC was used to provide predictions for both HLA class I and II molecules. These epitope predictions enable the selection of epitopes with a greater chance of stimulating T-cell responses. Tools in bioinformatics, such as NetMHC and IEDB, have greatly hastened the selection process of suitable epitopes with their predictive models of epitope binding to HLA molecules. Such tools assist the researcher in choosing epitopes that hold a high potential for inducing adequate immune responses and inhibiting tumor growth. The primary criterion for selection was a high binding affinity to HLA class II (HLA-DRB1_01:01) and class I (HLA-A11:01), with strong binders categorized at a binding affinity threshold of 2.0 nM and weak binders at 0.5 nM [19,20].

These epitopes must have low allergenicity and toxicity while demonstrating immunogenic solid potential to be suitable for therapeutic use. Antigenicity was determined by the Immunomedicine Group server (Immunomedicine Group) using the Kolaskar and Tongaonkar method, which has an approximate accuracy of 75% [20]. The ToxinPred server (ToxinPred) assessed the toxicity of the epitopes and provided predictions of their toxicity levels along with key physical and chemical properties such as hydrophobicity and pI charge [21]. Additionally, the allergenicity of the epitopes was predicted using the AllerTOP v. 2.0 server (AllerTOP v. 2.0), which utilizes the Automatic Cross-Covariance (ACC) transformation of protein sequences into equal-length vectors for protein sequence mining [22].

Ten epitopes with optimal properties were selected for the chimeric vaccine construct.

Chimeric Vaccine Design: The epitopes selected in the previous stage were used to design a chimeric structure, forming a multi-epitope DNA vaccine candidate. A flexible KK (lysine-lysine) linker was employed to connect these epitopes. Additionally, ProtParam (https://web.expasy. org/protparam/) and PepCalc (https://pepcalc.com/) were utilized to predict various physicochemical properties and amino acid composition of the vaccine construct.

ANTIGENpro (http://scratch.proteomics.ics.uci.edu/) was used to assess the antigenicity and allergenicity of the final vaccine. ANTIGENpro performs antigenicity predictions based on sequence analysis using an alignment-free and pathogen-independent approach. The AllerTOP v. 2.0 server (AllerTOP v. 2.0) was utilized for allergenicity assessment. TMHMM version 2.0 (http://www.cbs.dtu.dk/services/TMHMM/) was used For the final multi-epitope vaccine to identify potential transmembrane helices. TMHMM uses a hidden Markov model to predict the topology of membrane proteins and accurately distinguishes between soluble and membrane proteins [23].

Structural Analysis-Secondary and Tertiary Structure Prediction:

The Prabi server (https://prabi.ibcp.fr/htm/site/web/services/ secondaryStructurePrediction) uses the GOR IV method, known for its accuracy in predicting secondary structures, to predict the secondary structure of the protein vaccine [24]. This method predicts the secondary structure elements of the protein. For the three-dimensional structure prediction of the fused protein vaccine, the I-TASSER server (https://zhanglab.ccmb.med.umich.edu/ I-TASSER/) was utilized.

I-TASSER is renowned for generating high-quality and biologically active 3D structure models from an amino acid sequence by reconstructing three-dimensional structures from excised fragments extracted from threading templates. The models provided by I-TASSER are accompanied by a confidence score (C-score), which ranges from -5 to 2, indicating the reliability of the global structure, with higher scores representing more reliable models [25-27].

Tertiary Structure Refinement and Validation: The protocol involves a two-step process: optimizing the hydrogen bonding network and minimizing atomic energy by combining physics with knowledge-based force fields [28]. The GalaxyRefine server (https://galaxy.seoklab. org/cgi-bin/submit.cgi?type=REFINE) was utilized to enhance the quality of the previously obtained 3D model structure, refining it to represent its native conformation more accurately. Post-refinement validation of models is crucial to identify potential errors in initial structures and compare models before and after refinement [29].

For this purpose, three servers—RAMPAGE (http:// mordred.bioc.cam.ac.uk/~rapper/rampage.php), ProSA (https://prosa.services.came.sbg.ac.at/prosa.php), and ERRAT (https://servicesn.mbi.ucla.edu/ERRAT/) were employed in this study. The RAMPAGE server generated the Ramachandran plot, a tool used to evaluate the quality of protein structures derived from experimental methods such as X-ray crystallography, NMR, and cryo-EM. This plot distinguishes between suitable and poor-quality protein structures based on torsional angles in allowed and disallowed regions, respectively [30]. The ProSA server calculated a z-score to estimate the overall quality of the protein models. Z-scores falling outside the characteristic range of native proteins suggest potential structural inaccuracies [31]. ERRAT, on the other hand, is a tool specifically designed to evaluate crystallography-determined protein structures. It assesses specific atomic interactions to identify correct or incorrect regions within the protein structure [32].

Molecular Docking Study: Binding Affinity Evaluation: Binding Affinity Evaluation: Molecular docking was performed using the ClusPro 2.0 server (https://cluspro. org) between the final structure of the vaccine candidate and the cellular targets HLA-A11:01 (PDB ID: 1X7Q) and HLA-DRB1_01:01 (PDB ID: 1AQD) after preparing and energy minimizing each target protein.

Molecular Dynamics (MD) Simulation: Stability Assessment: The selected docking model was used as the MD simulation’s initial structure to verify the engineered vaccine binding stability to the HLA-DRB1_01:01 and HLA-A11:01 molecules. Gromacs 5.1.5 (Groningen MAchine for Chemical Simulations) software was utilized for this purpose. Input structures were prepared using the amber ff99SB force field, with necessary adjustments to ensure accurate hydrogen states for all histidine residues and define disulfide bonds within the protein. The surface charge of the structure was neutralized by adding sodium and chlorine ions. The protein was immersed in a layer of SPC water molecules at a distance of 10 angstroms. Energy minimization was performed using a 5000-step steepest descent method to eliminate van der Waals interactions and promote hydrogen bonding between water and complex molecules. The system temperature was gradually raised from 0 to 300 K over 100 picoseconds at constant volume, followed by equilibration for 100 picoseconds at constant pressure. MD simulation was conducted at 27°C (300 K) for 40 nanoseconds, calculating non-bonding interactions within a 10-angstrom distance using the Particle Mesh Ewald (PME) method. The SHAKE algorithm was employed to constrain hydrogen atom bonding. Simulation data were recorded at 0.4 picosecond intervals for subsequent analysis to accelerate the calculations.

Reverse Translation, Codon Optimization, and in Silico Cloning: Back translation of the amino acid sequence of the designed vaccine was conducted using the Gene infinity server (http://www.geneinfinity.org/sms/sms_ backtranslation.html). Codon adaptation is required to express a foreign gene in a host; therefore, optimization of the construct was assessed using the JCat service (http:// jcat.de/) by employing the commonly used E. coli K12 as the host. The codon adaptation index (CAI) value and GC content of the modified DNA were used to evaluate the sequence optimization. Finally, the in-silico cloning was performed by inserting the optimized DNA sequence into the pET30a (+) expression vector using the SnapGene v5.3.1 software.

In Vitro Expression Analysis

Plasmid Construction and Transfection Procedures: First, a chimeric gene was constructed and inserted into the pcDNA3.1+ expression vector. HEK293T cells, derived from human embryonic kidney cells, were then cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum (FBS), 100 units/mL penicillin, and 100 µg/mL streptomycin. These cells were maintained at 37°C in a humidified environment with 5% CO2.

A total of 2×105 cells were added to each well of a 6-well plate and allowed to adhere overnight to begin transfection. Transfections were performed using Lipofectamine 2000 reagent according to the manufacturer’s instructions after subcloning the chimeric gene into the pcDNA3.1+ vector.

Immunoblotting Analysis of Protein Expression: Forty-eight hours post-transfection, cells were harvested and washed with PBS solution. They were lysed using a Radioimmunoprecipitation Assay (RIPA) buffer containing protease inhibitors. The Bradford technique was employed to determine the protein concentration. Each sample, containing 20 µg of protein, was mixed with 4× Laemmli loading buffer and heated to 95°C for 5 minutes. The proteins were then separated on a 12% SDS-PAGE gel for 1.5 hours at 100V.

Following separation, the proteins were transferred onto a PVDF membrane using a semi-dry transfer device set at 20V for 45 minutes. The membrane was then stained with Ponceau S solution to confirm the efficiency of the transfer and observed using common imaging tools.

The PVDF membrane was blocked with 5% nonfat dry milk in Tris-buffered saline containing 0.1% Tween 20 (TBST) for 1 hour at room temperature to prevent nonspecific binding. The membrane was then incubated overnight at 4°C with a polyclonal anti-HPV16 E7 primary antibody (product # PA5-117383, Thermo Fisher Scientific) diluted 1:1000 according to the manufacturer’s instructions.

The membrane was washed three times with TBST before incubation with horseradish peroxidase (HRP)-coupled secondary antibodies diluted 1:5000 for 1 hour at room temperature. The membrane was washed thrice in TBST to remove any unbound secondary antibodies.

Cell-Based Immunogenicity (Cell-ELISA) – Preliminary Cytokine Readout

An optimized cell-based ELISA protocol was employed in our laboratory to evaluate the binding of antibodies raised by vaccination against HPV16 E6/E7 antigens expressed on the surface of the TC-1 cell line. The TC-1 cell line (a C57BL/6 mouse lung epithelial cell line transformed with HPV16 E6/E7 and activated H-ras oncogenes) was kindly provided by the Department of Medical Virology, Faculty of Medicine, Tarbiat Modares University, Tehran, Iran. Cells were maintained in RPMI-1640 medium (Gibco, USA) supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine, 1 mM sodium pyruvate, non-essential amino acids, and penicillin/streptomycin, at 37 °C in a humidified incubator with 5% CO?.

At approximately 90% confluency, TC-1 cells were harvested by trypsin digestion and resuspended in complete RPMI-1640. For coating, 200 µl of a cell suspension containing 3×104 cells was added into poly-D-lysine–coated 96-well plates (Nest, China) and cultured for 36 h under standard incubation conditions. After incubation, plates were washed three times with washing buffer (10 mM PBS containing 0.05% Tween-20). Cells were then fixed with 100 µl of 10% paraformaldehyde in PBS for 15 min at room temperature, followed by three additional washes. Blocking was performed with 10 mM PBS containing 0.05% Tween-20, 5% skimmed milk, and 2% normal goat serum for 1 h at 37°C.

We serially diluted serum samples from healthy vaccinated mice (group E), collected at day 7 after the second immunization (two-dose regimen), in blocking buffer and added them to the wells (100 µl per well, in duplicate) for 2 h at room temperature. Control wells included normal serum (negative control), conjugate-only wells (secondary antibody without primary), and cell-only wells (no antibody). After incubation, plates were washed three times with washing buffer and incubated with 100 µl of goat anti-mouse IgG conjugated to HRP (1/8000; Razirad, Iran) for 1 h at room temperature. Following five washes, 100 µl of TMB substrate was added to each well and incubated for 10 min before stopping the reaction with 100 µl of 2 N H?SO?. Optical density was measured at 450 nm with a reference wavelength of 570 nm using a microplate reader (BioTek, USA).

Statistics: Outcome measures included antigen-specific binding (mean ODcorr) across serum dilutions, area-under-curve (AUC) analysis, and signal-to-background (S/B) ratios relative to control sera. Normality of data distribution was assessed using the Shapiro–Wilk test. For comparisons between vaccinated (group E) and control groups, one-way ANOVA followed by Tukey’s post-hoc test was applied. In selected cases, unpaired two-tailed t-tests were used for pairwise comparisons. IgG isotype ratios (IgG2a/IgG1) were analyzed as exploratory endpoints. Data are presented as mean ± SEM from ≥3 biological replicates (mice), each tested in ≥2 technical replicates. Effect sizes and 95% confidence intervals (CI) were calculated, with significance set at α = 0.05.

Summary of Key Results

Our study aimed to develop a therapeutic vaccine against HPV16 by identifying and predicting immune-stimulating epitopes from the E6 and E7 proteins and the protein HSP110. The epitope prediction was performed computationally to ensure that such epitopes have high binding affinity to immune-activating molecules known as HLA, which are important for eliciting an immune response. These selected epitopes were further analyzed for their ability to induce a robust immune response against targeting HPV-infected cells specifically. The results demonstrated that the vaccine construct effectively stimulated the host’s immune system by inducing a robust T-cell response, suggesting its potential as a novel therapeutic strategy for treating cervical cancer caused by human papillomavirus infection.

In Silico Validation

Immunoinformatic Analysis And Epitope Prediction: T-cell epitope prediction and selection: Epitope prediction identifies small regions of proteins most likely to interact with the immune system. In this regard, we epitope-predicted from HPV16 proteins those that might induce a robust immune response by binding to HLA molecules. The NetMHC 4.0 and NetMHCII 2.3 servers predicted 16, 4, and 107 T-cell epitopes for the E6, E7, and HSP110 proteins. These predicted epitopes were screened for antigenicity, toxicity, and allergenicity. After analyzing the overlap between the epitopes, 3, 2, and 4 were finally selected for the E6, E7, and HSP110 proteins, respectively (Figure 2).

Figure 2: A) Schematic diagram of the designed chimeric protein. The blue, red, and green regions indicate T-cell epitopes of E6, E7, and HSP110, which are joined together by the KK linker (yellow regions). B) Secondary structure of the multi-epitope vaccine. The vaccine structure comprises 23.58% alpha helix, 17.92% extended strand, and 58.49% random coil. C) Predicting the topological properties of a protein derived from recombinant gene expression. Based on the graph above, all amino acids are most likely expressed in extracellular regions.

Chimeric Vaccine Design: Construct Composition: The final vaccine comprised epitopes from E680-90, 47 57, 86-95, E748-57, 41-47, and HSP11087-93, 15-22, 392 400, 196-203, 363-371 linked to the KK linker. A schematic of the multi-epitope vaccine was generated using the Illustrator for Biological Sequences (IBS) Ver 1.0 server (Figure 2).

Structural Analysis: Secondary and tertiary structure prediction: The Prabi server predicted the secondary structure of the protein vaccine, revealing it to consist of 28.83% alpha helix, 19.82% extended strand, and 51.35% random coil (Figure 2B). For the 3D structure prediction, the I-TASSER server generated five models of the vaccine construct based on ten threading templates (PDB Hit: 1yuwA, 4j8f, 3iucC, 4cj7A, 2v7yA, 5tky, 2khoA, 4j8f, 3iucC, and 4rtfA). The C-scores for these models were -1.86, -3.39, -4.18, -4.51, and -5, respectively. The C-score indicates confidence in the quality of the I-TASSER predicted models and typically falls within the range of [-5, 2]. A higher C-score signifies a higher confidence level in the model. Consequently, the model with a C-score of -1.86 was chosen for refinement.

Prediction of Properties: The results of Protparam analysis showed a theoretical pI of 9.78, indicating the basic nature of the vaccine construct, with a net charge of 15.9 at pH 7. The multi-epitope vaccine showed good solubility in water (Table 1).

Table 1: The computation of various physical and chemical parameters for the entered protein sequence. The computed parameters include the molecular weight, theoretical pI, amino acid composition, atomic composition, extinction coefficient, estimated half-life, instability index, aliphatic index, and grand average of hydropathicity (GRAVY). All investigations showed that the designed protein is temperature stable, has a suitable half-life, and is hydrophilic.

ANTIGENpro predicted an antigenicity probability of 0.285, indicating potential antigenic properties. The AllerTOP analysis indicated that the proposed vaccine was not allergenic. According to data from the TMHMM server (Figure 2C), all amino acids resulting from recombinant gene expression were located outside the cell membrane.

Refinement and Validation of the Tertiary Structure: The 3Drefine server refined the protein model obtained in the previous step. Various criteria were used to rank the optimized models, including the 3Drefine score, GDT-TS, GDT-HA, RMSD, RWplus, and MolProbity. Based on these assessments, Refined Model 5 was selected and detailed in the first row of Table 2.

Table 2. The model refinement results. The refined model No.5 was selected based on the 3Drefine Score, GDT-TS, GDT-HA, RMSD, MolProbity, and RWPlus parameters.

The raw three-dimensional structure, the refined model of the multi-epitope vaccine, and the comparison of two primary and refined structures are visualized in Figure 3A. As illustrated in the figure, the structure is optimized in terms of the position of the atoms, resulting in decreased internal energy in the system. Additionally, the secondary structures in the model have been optimally modified.

Checking the geometric quality of the initial model and the refined model, in addition to the ProSA server (zplot and eplot charts), was evaluated and compared by the Ramaplot server (Figure 3B and 3C). ERRAT, on the other hand, is a tool specifically designed to evaluate crystallography-determined protein structures. It assesses specific atomic interactions to identify correct or incorrect regions within the protein structure (Figure 3D) (46). The investigation results indicated the structure’s improvement compared to the initial model.

Figure 3: (A) Protein modeling and refinement. (1) The Three-dimensional structure of the proposed vaccine was obtained from the I-TASSER server; (2) The final 3D model of the vaccine was refined by the 3Drefine server; (3) Comparison of two primary and refined structures; The initial structure is shown in red, and the refined structure is shown in green; (B) 3D protein model validation using the Ramachandran plot before and after refining. (1) In the initial model, 70%, 20%, and 10% of residues were positioned in favored, allowed, and outlier regions, respectively. (2) In the refined model, 92%, 7%, and 1% of residues were observed in the Core, allowed, and Disallowed regions, respectively. It shows the concentration of core areas in the allowed parts of the chart. (C) ProSA-web z-score plot for primary and refined models. (1) The z-score in the primary model was -4.13 (2), while this value was changed to -5.4 in the refined model. The overall quality factor index has increased significantly in the refined structure. (D) The ERRAT Plot displays the overall quality factor of both the original and refined models. (1) The initial model demonstrated an overall quality factor of 29.592, (2) notably increased to 83.673 after refinement.

Molecular Docking and Dynamics: Binding Affinity Analysis: The docking of the refined vaccine candidate structure, achieved using the GalaxyRefine web server, was performed with the cellular targets HLA-A11:01 (PDB ID: 1X7Q) and HLA-DRB1_01:01 (PDB ID: 1AQD). The molecular docking study, therefore, is an in silico method used to predict the affinity of these epitopes toward the immune molecules, HLA class I/II. This interaction will ensure that the immune system effectively recognizes the designed vaccine, eliciting an appropriate response. Before docking, each target protein was prepared and underwent energy minimization. ClusPro 2.0 was employed to analyze the binding sites and relative binding strengths of the proteins, presenting results within clusters according to their binding energy. The model with the largest cluster size and the lowest energy score was selected. Consequently, these two models were identified as the best-docked complexes, demonstrating the interaction between the vaccine candidate and the HLA proteins (Figure 4).

Figure 4: Molecular docking of the selected epitopes from chimeric vaccine with HLA-A11:01 and HLA-DRB1_01:01 molecules. Linear peptide sequences of vaccine candidates from HPV16 E6, E7, and HSP110 predicted for strong binding affinity to HLA molecules were docked to the binding pockets of HLA-A11:01 (PDB ID:1X7Q) and HLA-DRB1_01:01(PDB ID:1AQD). The vaccine candidate is depicted in blue, while the HLA proteins are shown in green. Additionally, the amino acids involved in the binding have been identified through staining techniques related to the vaccine and proteins.

MD Simulation: MD simulations were conducted using the GROMACS 2019.6 software with the G43a1 force field and SPC water model. The system was neutralized and achieved a physiological ionic strength of 0.14 M by adding Na+ and Cl− ions. Initial energy minimization was performed using the steepest descent method with specific cut-offs for short-range interactions. Equilibration was conducted in the NVT and NPT ensembles at 300 K for 500 and 1000 ps using the Berendsen thermostat and barostat. A 20 ns MD production run was performed at 310 K and 1 atm, with temperature and pressure coupling via the Nosé-Hoover thermostat and Parrinello-Rahman barostat.The leap-frog algorithm was used for motion integration, and the LINCS algorithm maintained bond constraints while the Particle Mesh Ewald method handled long-range electrostatic interactions.

The root means square deviation (RMSD) analysis indicated initial increases followed by stability in the vaccine-receptor complexes, with average RMSD values of 0.91 ± 0.04 nm and 0.53 ± 0.05 nm for HLA-DRB1_01:01 vaccine and HLA-A11:01-vaccine complexes, respectively. Root means square fluctuation (RMSF) analysis showed higher flexibility in certain chains, particularly in HLA-DRB1_01:01 regions 1–380 and HLA-A11:01 chain B. The radius of gyration (Rg) values suggested greater compactness in the HLA-A11:01-vaccine complex, attributed to its smaller protein size.

Hydrogen bond analysis revealed increasing interaction stability, with average hydrogen bonds of 16.41 ± 5.24 for HLA-DRB1_01:01-vaccine and 8.39 ± 2.58 for HLA-A11:01-vaccine. Solvent accessible surface area (SASA) analysis further indicated stable interactions, with values of 705.85 ± 23.59 nm² and 233.37 ± 6.19 nm² for HLA-DRB1_01:01 and HLA-A11:01, respectively (Figure 5 and 6).

Figure 5: (A) The root-mean-square deviation (RMSD) plot of the vaccine constructs. (B) RMSF) plot of the HLA-DRB1_01:01-vaccine. (C) The root-mean-square fluctuation (RMSF) plot of the HLA-A11:01-vaccine complex. (D) The designed vaccine’s root-mean-square fluctuation (RMSF) plot into the vaccine complexes. (E) Radius of Gyration (Rg) plot of the vaccine systems. (F) SASA plot of the vaccine complexes. (G) The number of hydrogen bonds during simulation.

Figure 6: Population coverage analysis of the selected T-cell epitopes designed in this study for the Iranian population. The graphs illustrate the predicted population coverage for MHC class I epitopes, MHC class II epitopes, and the combined class I and class II epitopes. The analysis showed an estimated coverage of approximately 59.1% for MHC class I and 28.12% for MHC class II, resulting in an overall population coverage of about 70.54% in the Iranian population.

HLA-DRB1_01:01-vaccine complex exhibits a higher number of hydrogen bonds and a higher SASA value, indicating strong, relatively stable interactions but with more surface area exposed to the solvent. The HLA-A11:01-vaccine complex exhibits fewer hydrogen bonds but has a lower SASA value, suggesting that the interactions, while fewer, are tighter and potentially more stable due to less solvent exposure.

Revers translation, codon optimization, and in silico cloning: Back translation of the vaccine construct was conducted using the Gene infinity server, and adjusting the codons according to the host was assessed by the JCat service to improve the translation efficiency of the vaccine (Figure 7A). The modified nucleotide sequence has a GC content of 42.76% and a CAI value of 1.0 in E. coli, respectively. To insert the vaccine sequence into the pET30a (+) vector, we employed the BamHI and HindIII restriction sites as the start and end of the optimized gene, respectively (Figure 7B). The developed vaccine sequence was cloned into the pET30a (+) vector using the SnapGene software.

Validation of Cloning and Protein Expression

Cloning Confirmation: The pET30a (+) chimeric gene was commercially synthesized and then successfully subcloned into the pcDNA3.1+ vector between the BamHI and HindIII restriction sites and transfected into HEK293 cells (Figure 7C).

Protein Expression Analysis: Western blot analysis confirmed the expression of HPV-specific epitopes in transfected cells using HPV E7 polyclonal antibody. The in-house-generated HPV E7 polyclonal antibody utilized in this study can recognize multiple epitopes across the full-length E7 protein. In this way, the antibody can be guaranteed to detect both the intact E7 protein and the two E7-derived peptides incorporated into the chimeric vaccine construct, thus providing comprehensive detection in the Western blot analysis.

Specific bands corresponding to the chimeric gene were detected, confirming the successful expression of the designed construct (Figure 7D). The presence of distinct bands matching chimeric protein proves that the construct is accurately produced in a eukaryotic system.

Figure 7: A) In silico restriction cloning of the designed vaccine sequence into the pET30a (+) expression vector. The red region represents the vaccine coding gene, and the black circle represents the vector backbone. B) Codon adaptation of the multi-epitope vaccine to E. coli K12 strain. C) Schematic representation of the pCDNA3.1(+)-E6/E7/HSP110 plasmid. D) The western blot analysis of the chimeric protein. Lanes 1 and 2: chimeric protein, L: protein ladder. The Western blot bands validated the presence and size of the expressed protein (12.3 kD). This antibody recognizes multiple epitopes in the complete E7 protein to ensure the comprehensive detection of antigens from E7, as the complete protein and peptide fragments present in this chimeric vaccine construct.

Cell-ELISA Readout

Serum samples from healthy vaccinated mice (group E), collected at day 7 post-second immunization, were analyzed for antigen-specific antibody binding using a cell-based ELISA on TC-1 cells expressing HPV16 E6/ E7 antigens. Optical densities (ODs) were background=corrected (ODcorr) by subtracting the mean signal of antibody-free control wells (groups F–H).

Vaccinated sera (group E) displayed significantly higher ODcorr values compared to control sera across all dilutions (1/50–1/800), indicating robust binding of vaccine induced antibodies to TC-1 cells. The primary endpoint— mean ODcorr of group E versus controls—showed a clear antigen-specific signal. Secondary analyses confirmed this finding: (i) the area-under-curve (AUC) across serial dilutions was markedly elevated in group E, (ii) the signal to-background (S/B) ratio was substantially greater than 1, and (iii) preliminary IgG isotyping suggested a shift toward IgG2a dominance, consistent with a Th1-skewed immune response.

Together, these results provide functional evidence that the designed DNA vaccine elicits circulating antibodies capable of recognizing HPV16-derived antigens on tumor derived TC-1 cells, thereby supporting its immunogenic potential in vivo.

In this paper, a new therapeutic DNA vaccine is developed that targets the HPV16 E6, E7, and HSP110 proteins as treatment methods for cervical cancer. The stability and efficacy of the vaccine were approved by advanced immunoinformatics techniques in this study. It is inferred from this work that it might produce an effective immune response against the human body. The prophylactic vaccines already developed are useless for the previously infected person. Our therapeutic vaccine will provoke an immune response against selective killing of the HPV-infected cells. Epitope prediction, molecular docking, and dynamics simulations have been done to validate vaccine design in silico. The study, therefore, paves the way for its further experimental development.

Indeed, previous studies have identified HPV16 E6 and E7 as critical targets in cervical cancer immunotherapy based on their continued expression within the malignant cells. In the working vaccine, these antigens are alloyed with HSP110, a heat shock protein that enhances immune response and possibly includes a more potent treatment than existing prophylactic vaccines. Incorporating HSP110 in the vaccine design significantly enhances antigen presentation, which is crucial for activating innate and adaptive immune responses. This may render the vaccine even more effective than any existing prophylactic option in infected individuals. The multi-epitope approach also allows for a wide immune response against more regions of the oncoproteins, offering greater efficiency in tumor clearance. This is more robust than previous methods, which typically rely on E6 and E7, as proof has been provided that adding HSP110 enhances the response by increasing immunogenicity.

Moreover, our study finds strength in the output of similar immunotherapeutic analyses that have proved efficient in tumor regression models. These studies give credence to the underlying theoretical basis for our designed vaccine, as they have shown high advantages in targeting or eliminating the tumors emanating from HPV. The vaccine is anticipated to enhance immunogenicity over existing preventive vaccines, as evidenced by its ability to induce a robust T-cell-mediated response, which is crucial for combating persistent infections and established cancers. This study designed a novel chimeric structure combining T-cell epitopes from E6, E7, and HSP110 as a multi-epitope vaccine.

Previous studies by Jabbar et al., and Yao et al., predicted antigenic peptides and cytotoxic T lymphocyte epitopes, respectively, for E6 and E7 proteins using immunoformatic methods and immune epitope databases [33,34]. However, limited efforts have incorporated bioinformatics analyses, structural predictions, or MHC molecule interactions for chimeric structures containing E6E7 epitopes. For instance, de Oliveira et al. reported a vaccine’s efficacy in murine tumor models but lacked comprehensive bioinformatics evaluations of the vaccine structure [15].

Bioinformatics analyses were pivotal in epitope selection, where high-ranking epitopes were chosen based on antigenicity, toxicity, and allergenicity. Incorporating HSP110 epitopes aimed to enhance the vaccine’s potency,given reports of HSP110-derived peptides inducing specific CTLs with potential anti-tumor activity [35]. HSP110 is a classical chaperone and is also involved in boosting antigen presentation. HSP110-derived peptides can be recognized and presented to immune cells (CD4+ helper T-cells) by antigen-presenting cells, and HSP110 can aid in the folding and presentation of these peptides, thereby enhancing the immune response. HSPs, like HSP110, HSP70, and gp96, have been shown to boost immune responses targeted at specific antigens. These proteins work as enhancers by aiding in the presentation of antigens and enhancing the body’s overall immune response and cellular immunity [36]. Fusion proteins that merge HPV antigens with HSPs have displayed higher levels of immunogenicity and effectiveness in treatment in different experimental models [37]. Although they are smaller than the whole protein, it has been shown that the peptides generated from HSP110 retain the competence of presenting epitopes via APCs that may effectively activate T-cells. Within the heat shock proteins family, the HSP110 proteins family is well established as playing a role in enhancing antigen presentation and increasing the immunorecognition of tumor antigens [14]. Our strategy of choosing the most immunogenic HSP110 regions ensures that the vaccine design can still benefit from the immune enhancing properties of HSP110, focusing this time on the cellular immune response.

The selected epitopes were amalgamated using KK linkers to construct the vaccine. KK linkers are effective spacers in vaccine designs, aiding proteasome processing and presentation of MHC -MHC-restricted antigens [38].

The structural analysis of the vaccine construct involved secondary and tertiary structure prediction, refinement, and validation. For the secondary structural analysis, properties included molecular weight, isoelectric point, net charge at pH 7, estimated solubility in water, estimated half-life in mammalian reticulocytes, instability index, aliphatic index, and grand average of hydropathicity (GRAVY) were evaluated. Molecular weight influences transport, binding, and enzymatic reactions. The isoelectric point (pI), at which an amino acid or peptide carries no net charge, is crucial in understanding protein behavior across different pH environments. Consider the net charge at pH 7, which affects interactions with other molecules.

Additionally, estimated solubility in water influences transport and distribution within cells. Understanding the estimated half-life in mammalian reticulocytes is vital for drug design and protein turnover. The instability index predicts protein stability, while the aliphatic index correlates with thermostability. Lastly, the grand average of hydropathicity (GRAVY), reflecting hydrophobicity based on amino acid composition, informs protein behavior. These properties, collectively, empower researchers to design better drugs, predict protein behavior, and unravel intricate biological processes. Assessment of the designed vaccine’s physicochemical properties indicated favorable attributes [39-41]. The vaccine exhibited a molecular weight of 12.3 kDa, a basic nature with a theoretical pI of 9.78, good solubility in water, and a prolonged half-life of over 30 hours in yeast, making it conducive for immunotherapy. Additionally, the absence of transmembrane helices in the vaccine’s structure indicated ease of expression in engineered systems. Notably, the vaccine displayed antigenicity and non-allergenicity.

Using the I-TASSER server provided high-quality 3D models with a satisfactory C-score, indicative of the model’s reliability. The subsequent refinement using the 3Drefine server further optimized the structure, enhancing its stability as confirmed by various validation tools, including RAMPAGE, ProSA, and ERRAT. These analyses demonstrated significant improvements in the structural quality of the vaccine, which is essential for its potential efficacy in vivo.

Molecular docking studies revealed binding solid affinities between the vaccine construct and HLA class I (HLA-A11:01) and class II (HLA-DRB1_01:01) molecules. The docking models demonstrated low-energy scores, suggesting robust interactions crucial for eliciting a strong immune response. Furthermore, molecular dynamics (MD) simulations confirmed the stability of these interactions over time. The RMSD, RMSF, and hydrogen bond analyses indicated stable and compact interactions, particularly in the HLA-A11:01-vaccine complex, which exhibited tighter and potentially more stable interactions than the HLA DRB1_01:01-vaccine complex.

Today, computational immunology has revolutionized vaccine development. Researchers can swiftly identify promising vaccine candidates by predicting immunogenic epitopes, optimizing gene properties, and assessing stability through molecular dynamics simulations. Some proposed vaccines, refined using computational tools, now await comprehensive in vitro and in vivo evaluations [42 46].

Commercial vaccines protect against HPV, but those who are unvaccinated or already infected with the virus remain at risk of developing precancerous lesions and cancer from high-risk strains such as HPV16 and HPV 18 [47]. Developing therapeutic HPV vaccines is crucial to treating existing infections and preventing cancer progression. Although vaccine development is typically lengthy and costly, reverse vaccinology and vaccinomics have been developed to accelerate this process [48,49].

Reverse vaccinology, which combines immunogenomics and bioinformatics, offers significant benefits over traditional methods. It helps reduce the time and costs of vaccine development, making it a valuable approach [50].This method has been essential in predicting epitopes for developing multi-epitope vaccines for various organisms, including Hepatitis B Virus, Onchocerca volvulus, Klebsiella pneumonia, Mycobacterium tuberculosis, and Helicobacter pylori.

Previous therapeutic vaccines have targeted the HPV16 E6 and E7 proteins, but such studies have lacked an integrated approach using epitope prediction by bioinformatics with potent enhancers of the immune response, namely, HSP110. Further investigation of epitope prediction by advanced computational tools, interaction with HLA molecules, the stability of the latter interaction through molecular docking, and dynamics simulation is presented in this paper. These techniques give way to a more robust and reliable vaccine construct with higher precision and effectiveness than traditional methods. In this study, we have improved the originality of our analysis by presenting a new approach for epitope prediction that uses sophisticated machine learning techniques. This novel approach has shown enhanced precision compared to conventional methods.

Furthermore, we have performed a novel technique for improving the accuracy of the 3D structure by incorporating it with our epitope prediction algorithm. This technology offers very dependable structural predictions that are essential for vaccine development. Future research should build on these results by conducting thorough in vivo and clinical evaluations, improving delivery systems, and exploring combination therapies [51-54].

This study has certain limitations that warrant consideration. Most of the research was primarily conducted using in silico methods, including epitope prediction, molecular docking, and molecular dynamics simulations. While these computational approaches are crucial for efficiently designing the vaccine and reducing initial costs, they require further experimental validation in laboratory and clinical settings to fully assess its efficacy and safety. Results derived from computational models may vary in practical applications, highlighting the need for continued validation.

Additionally, the vaccine has not yet been evaluated in clinical trials. Although the computational and predictive data are promising, the actual immune response elicited by the vaccine in human subjects remains to be determined. Preclinical and clinical studies will confirm these results and ensure the vaccine’s safety and effectiveness.

Another limitation is that animal model testing was not included in this phase. Animal studies could provide crucial information on how the vaccine interacts in a living system and whether it effectively targets HPV-infected cells. Future work will focus on preclinical testing in relevant animal models to bridge the gap between computational findings and clinical outcomes.

Despite these limitations, the study presents notable strengths. Advanced bioinformatics tools such as epitope prediction and molecular docking enabled the selection of highly promising epitopes, potentially enhancing the specificity and efficacy of the immune response against HPV. Moreover, including HSP110 as an immune adjuvant is an innovative approach aiming to significantly boost innate and adaptive immune responses, setting our vaccine apart from other candidates that target only E6 and E7 proteins.

In summary, while experimental and clinical validations are still required, the promising in silico findings and the innovative combination of HPV oncoproteins with HSP110 provide a strong foundation for future studies. The designed vaccine represents a potentially effective therapeutic option, especially for individuals beyond the reach of existing prophylactic vaccines, marking a meaningful advancement in the field of HPV-related cervical cancer therapy.

Based on the data from the study, this new approach may represent a novel and promising method for designing a therapeutic vaccine against cervical cancer. Immunoinformatics analysis can identify potent epitopes capable of inducing targeted and robust immune responses, making this strategy particularly effective. Although further confirmation through experimental validation and clinical trials is needed, our study suggests preliminary evidence to guide the discovery of new therapy methods for cervical cancer. The integration of these two fields is attractive due to the promise of developing new cancer treatments and ending some existing ones.

Author Contributions

• Maryam Fazeli: Conceptualization, Methodology, Investigation, Validation, Writing – Original Draft

• Babak Mostafazadeh: Validation, Writing – Review & Editing

• Hesamoddin Ahmadi Afzadi: Data Curation, Formal Analysis, Software, Visualization

• Zahra Ziafati Kafi: Methodology, Validation, Writing – Review & Editing

• Bahram Zarmehri: Validation, Writing – Review & Editing

• Behzad Pourhossein: Resources, Funding Acquisition, Conceptualization, Writing – Review & Editing

• Ramin Sarrami Forooshani: Supervision, Conceptualization, Validation, Writing – Review & Editing

Availability of Data and Materials

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

1. zur Hausen H. Papillomaviruses and cancer: from basic studies to clinical application. Nat Rev Cancer. 2002; 2: 342-350.
2. de Martel C, Plummer M, Vignat J, Franceschi S. Worldwide burden of cancer attributable to HPV by site, country and HPV type. Int J Cancer. 2017; 141: 664-670.
3. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018; 68: 394-424.
4. Hancock G, Hellner K, Dorrell L. Therapeutic HPV vaccines. Best Pract Res Clin Obstet Gynaecol. 2018; 47: 59-72.
5. Hung CF, Ma B, Monie A, Tsen SW, Wu TC. Therapeutic human papillomavirus vaccines: current clinical trials and future directions. Expert Opin Biol Ther. 2008; 8: 421-439.
6. Molano M, Van den Brule A, Plummer M, Weiderpass E, Posso H, Arslan A, et al. Determinants of clearance of human papillomavirus infections in Colombian women with normal cytology: a population-based, 5-year follow-up study. Am J Epidemiol. 2003; 158: 486-494.
7. Mo Y, Ma J, Zhang H, Shen J, Chen J, Hong J, et al. Prophylactic and Therapeutic HPV Vaccines: Current Scenario and Perspectives. Front Cell Infect Microbiol. 2022; 12: 909223.
8. Li Y, Subjeck J, Yang G, Repasky E, Wang XY. Generation of anti-tumor immunity using mammalian heat shock protein 70 DNA vaccines for cancer immunotherapy. Vaccine. 2006; 24: 5360-5370.
9. Suhrbier A. Multi-epitope DNA vaccines. Immunol Cell Biol. 1997; 75: 402-408.
10. Rueckert C, Guzmán CA. Vaccines: from empirical development to rational design. PLoS Pathog. 2012; 8: e1003001.
11. Hickman-Miller HD, Hildebrand WH. The immune response under stress: the role of HSP-derived peptides. Trends Immunol. 2004; 25: 427-433.
12. Farzanehpour M, Soleimanjahi H, Hassan ZM, Amanzadeh A, Ghaemi A, Fazeli M. HSP70 modified response against HPV based tumor. Eur Rev Med Pharmacol Sci. 2013; 17: 228-234.
13. Segal BH, Wang XY, Dennis CG, Youn R, Repasky EA, Manjili MH, et 

    al Heat shock proteins as vaccine adjuvants in infections and cancer. 2006; 11: 534-540.

14. Bolhassani A, Rafati S. Mini-chaperones: potential immuno-stimulators in vaccine design. Hum Vaccin Immunother. 2013; 9: 153-161.
15. de Oliveira LM, Morale MG, Chaves AA, Cavalher AM, Lopes AS, Diniz Mde O, et al. Design, Immune Responses and Anti-Tumor Potential of an HPV16 E6E7 Multi-Epitope Vaccine. PLoS One. 2015; 10: e0138686.
16. Gao J, Fan L, Ma W, Xiao HJM, immunology: Synergistic antitumor effect of a human papillomavirus DNA vaccine harboring E6E7 fusion gene and vascular endothelial growth factor receptor 2 gene. 2016; 60: 626-633.
17. Vici P, Pizzuti L, Mariani L, Zampa G, Santini D, Di Lauro L, et al. Targeting immune response with therapeutic vaccines in premalignant lesions and cervical cancer: hope or reality from clinical studies. Expert Rev Vaccines. 2016; 15: 1327-1336.
18. Yang A, Farmer E, Wu TC, Hung CF. Perspectives for therapeutic HPV vaccine development. J Biomed Sci. 2016; 23: 75.
19. Andreatta M, Nielsen M. Gapped sequence alignment using artificial neural networks: application to the MHC class I system. Bioinformatics. 2016; 32: 511-517.
20. Nielsen M, Lundegaard C, Worning P, Lauemøller SL, Lamberth K, Buus S, et al. Reliable prediction of T-cell epitopes using neural networks with novel sequence representations. Protein Sci. 2003; 12: 1007-1017.
21. Gupta S, Kapoor P, Chaudhary K, Gautam A, Kumar R, Consortium OSDD, Raghava GPJPo: In silico approach for predicting toxicity of peptides and proteins. 2013; 8: e73957.
22. Wold S, Jonsson J, Sjörström M, Sandberg M, Rännar SJACA: DNA and peptide sequences and chemical processes multivariately modelled by principal component analysis and partial least-squares projections to latent structures. 1993; 277: 239-253.
23. Magnan CN, Zeller M, Kayala MA, Vigil A, Randall A, Felgner PL, Baldi P. High-throughput prediction of protein antigenicity using protein microarray data. Bioinformatics. 2010; 26: 2936-2943.
24. Krogh A, Larsson B, von Heijne G, Sonnhammer EL. Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J Mol Biol. 2001; 305: 567-580.
25. Roy A, Kucukural A, Zhang Y. I-TASSER: a unified platform for automated protein structure and function prediction. Nat Protoc. 2010; 5: 725-738.
26. Yang J, Yan R, Roy A, Xu D, Poisson J, Zhang Y. The I-TASSER Suite: protein structure and function prediction. Nat Methods. 2015; 12: 7-8.
27. Zhang Y. I-TASSER server for protein 3D structure prediction. BMC Bioinformatics. 2008; 9: 40.
28. Kalisman N, Levi A, Maximova T, Reshef D, Zafriri-Lynn S, Gleyzer Y, et al. MESHI: a new library of Java classes for molecular modeling. Bioinformatics. 2005; 21: 3931-3932.
29. Nezafat N, Karimi Z, Eslami M, Mohkam M, Zandian S, Ghasemi Y. Designing an efficient multi-epitope peptide vaccine against Vibrio cholerae via combined immunoinformatics and protein interaction based approaches. Comput Biol Chem. 2016; 62: 82-95.
30. Kumar P, Arya AJPRTF: Ramachandran Plot: A simplified approach.2018: 1-7.
31. Wiederstein M, Sippl MJ. ProSA-web: interactive web service for the 

    recognition of errors in three-dimensional structures of proteins. Nucleic Acids Res. 2007; 35: W407-W410.

32. Colovos C, Yeates TO. Verification of protein structures: patterns of nonbonded atomic interactions. Protein Sci. 1993; 2: 1511-1519.
33. Jabbar B, Rafique S, Salo-Ahen OMH, Ali A, Munir M, Idrees M, et al. Antigenic Peptide Prediction From E6 and E7 Oncoproteins of HPV Types 16 and 18 for Therapeutic Vaccine Design Using Immunoinformatics and MD Simulation Analysis. Front Immunol. 2018; 9: 3000.
34. Yao Y, Huang W, Yang X, Sun W, Liu X, Cun W, et al. HPV-16 E6 and E7 protein T cell epitopes prediction analysis based on distributions of HLA-A loci across populations: an in silico approach. Vaccine. 2013; 31: 2289-2294.
35. Ren F, Xu Y, Mao L, Ou R, Ding Z, Zhang X, et al. Heat shock protein 110 improves the antitumor effects of the cytotoxic T lymphocyte epitope E7(49-57) in mice. Cancer Biol Ther. 2010; 9: 134-141.
36. Bolhassani A, Zahedifard F, Taghikhani M, Rafati S. Enhanced immunogenicity of HPV16E7 accompanied by Gp96 as an adjuvant in two vaccination strategies. Vaccine. 2008; 26: 3362-3370.
37. Sveshnikov P, Solopova O, Kurbatova E, IuS S, Baranovski? P, Ul’ianov A, et al. Heat-shock protein enhances immunogenicity of E7 oncoprotein of human papillomavirus included in chimeric construction (E7 HPV-HSP70). 2010; 3: 36-39.
38. Nosrati M, Hajizade A, Nazarian S, Amani J, Namvar Vansofla A, Tarverdizadeh Y. Designing a multi-epitope vaccine for cross-protection against Shigella spp: An immunoinformatics and structural vaccinology study. Mol Immunol. 2019; 116: 106-116.
39. Barh D, Barve N, Gupta K, Chandra S, Jain N, Tiwari S, et al. Exoproteome and secretome derived broad spectrum novel drug and vaccine candidates in Vibrio cholerae targeted by Piper betel derived compounds. PLoS One. 2013; 8: e52773.
40. Gasteiger E, Hoogland C, Gattiker A, Duvaud Se, Wilkins MR, Appel RD, et al. Protein identification and analysis tools on the ExPASy server. Springer; 2005.
41. Ali M, Pandey RK, Khatoon N, Narula A, Mishra A, Prajapati VK. Exploring dengue genome to construct a multi-epitope based subunit vaccine by utilizing immunoinformatics approach to battle against dengue infection. Sci Rep. 2017; 7: 9232.
42. Yang J, Zhang Y. I-TASSER server: new development for protein structure and function predictions. Nucleic Acids Res. 2015; 43: W174-W181.
43. Zhang C, Freddolino L, Zhang Y. COFACTOR: improved protein function prediction by combining structure, sequence and protein-protein interaction information. Nucleic Acids Res. 2017; 45: W291-W299.
44. Carbone A, Zinovyev A, Képès F. Codon adaptation index as a measure of dominating codon bias. Bioinformatics. 2003; 19: 2005-2015.
45. Puigbò P, Guzmán E, Romeu A, Garcia-Vallvé S. OPTIMIZER: a web server for optimizing the codon usage of DNA sequences. Nucleic Acids Res. 2007; 35: W126-W131.
46. Darabi P, Galehdari H, Khatami SR, Shahbazian N, Shafeei M, Jalali A, et al. Codon optimization, cloning and expression of the human Leukemia Inhibitory Factor (hLIF) in E. coli. 2013; 11: 47-53.
47. Burd EM. Human papillomavirus and cervical cancer. Clin Microbiol Rev. 2003; 16: 1-17.
48. Plotkin S. History of vaccination. Proc Natl Acad Sci U S A. 2014; 111: 12283-12287.
49. Poland GA, Ovsyannikova IG, Jacobson RM. Application of pharmacogenomics to vaccines. Pharmacogenomics. 2009; 10: 837-852.
50. Rappuoli R. Reverse vaccinology. Curr Opin Microbiol. 2000; 3: 445-450.
51. Zheng J, Lin X, Wang X, Zheng L, Lan S, Jin S, et al. In Silico Analysis of Epitope-Based Vaccine Candidates against Hepatitis B Virus Polymerase Protein. Viruses. 2017; 9: 112.
52. Dar HA, Zaheer T, Shehroz M, Ullah N, Naz K, Muhammad SA, et al. Immunoinformatics-Aided Design and Evaluation of a Potential Multi-Epitope Vaccine against Klebsiella Pneumoniae. Vaccines (Basel). 2019; 7: 88.
53. Moradi J, Tabrizi M, Izad M, Mosavari N, Feizabadi MMJJJoM: Designing a novel multi-epitope DNA-based vaccine against tuberculosis: in silico approach. 2017; 10.
54. Urrutia-Baca VH, Gomez-Flores R, De La Garza-Ramos MA, Tamez-Guerra P, Lucio-Sauceda DG, Rodríguez-Padilla MC. Immunoinformatics Approach to Design a Novel Epitope-Based Oral Vaccine Against Helicobacter pylori. J Comput Biol. 2019; 26: 1177-1190.