Toll-Like receptor 3 (TLR3) agonists in a multi-peptide vaccine for TFDP3 expressing cancers

doi:10.21203/rs.3.rs-5321374/v1

The increase in cancer incidence and mortality worldwide has demonstrated the need for investment in more effective anti-tumor therapies. Given the complexity of the mechanisms that lead to resistance to anti-tumor treatments, target therapies are promising approaches. Cancer testicular antigens (CTAs) are therapeutic targets with the potential to be explored, as they are not expressed in normal cells and are expressed in tumor cells, as is the case with TFDP3, expressed in triple-negative breast cancer, prostate cancer, childhood T-cell lymphoblastic leukemia and hepatocellular carcinoma. The objective proposed in this work is the in silicoprediction of a multi-epitope tumor antigen vaccine candidate from TFDP3. The epitopes were screened using immunoinformatics tools that identified the antigenic epitopes that interacted with B lymphocytes, CD4+, T lymphocytes, and CD8+ T lymphocytes. The population coverage of the epitopes on CD4+ T lymphocytes and CD8+ T lymphocytes was then assessed. From the epitopes of B lymphocytes, CD4+ T lymphocytes, and CD8+ T lymphocytes, 3 epitopes from each were selected to make up the multi-epitope vaccine determined by antigenicity, allergenicity, toxicity, IFN-γ induction, and population coverage. In addition to the epitopes, the vaccine was made up of an adjuvant and ligands that ensured certain properties of the epitopes, their processing in MHC class I biosynthesis, and post-translational modifications. The vaccine's homology with other proteins was assessed using the NCBI BLASTp server. The physicochemical parameters, antigenicity, allergenicity, and toxicity were then evaluated. The secondary structure and tertiary structure were determined using servers that use neural networks, as well as the quality parameters associated with the structure. In the tertiary structure, the linear and discontinuous epitopes of B lymphocytes were determined using the IEDB server. From there, the interaction by molecular docking with Toll-like receptors and molecular dynamics was evaluated to assess the stability of the multi-epitope vaccine in a biological system. Finally, the in silico assessment of the possibility of cloning the multi-epitope vaccine and its immune response after 1 and 3 successive administrations was also evaluated. Epitopes that interact with antigenic, non-allergenic, and non-toxic B lymphocytes, CD4+ T lymphocytes, and CD8+ T lymphocytes were identified. About CD4+ T lymphocytes, 4 epitopes, as well as being antigenic, non-allergenic, and non-toxic, are inducers of IFN-γ. In the population coverage, the MHC class I and MHC class II epitopes had 93.55% coverage worldwide. The multi-epitope vaccine has biologically favorable physicochemical parameters, low homology with human proteins, secondary and tertiary conformation compatible with native protein structures. It also has interactions with TLR-2 and TLR-3, with TLR-3 being the interaction that in a biological system guarantees the greatest stability of the multi-epitope vaccine. In addition, in silico analyses have shown that the multi-epitope vaccine can be cloned and develop a more robust and prolonged immune response when submitted to 3 administrations. Therefore, the multi-epitope vaccine designed from the testicular cancer antigen TFDP3 showed in silico several promising biological properties and responses so that in vitro and in vivo studies can be invested and the future application of this vaccine in the treatment of cancer types that express this CTA.

cancer

immunoinformatics

epitope prediction

peptide vaccine.

The World Health Organization's (WHO) epidemiological data has estimated an increase in cancer incidence and mortality worldwide (WHO, 2022). This projection has associated more with epigenomic factors than inheritable genomic ones. Given this scenario, several studies have advanced in the development of more effective anti-tumor therapies that include target therapies, i.e. those directed at a particular molecule that is involved in processes that stimulate carcinogenesis or lead to resistance to therapies.

The tumor microenvironment is made up of cellular and acellular components that play important roles in the process of carcinogenesis, such as stimulating angiogenesis, inhibiting the recognition of immune cells and promoting favorable conditions for the development of tumor cells in locations other than the primary focus (metastases) (Casey et al., 2015). Both the existence of tumor cell subclones and the characteristics of the tumor microenvironment are factors that contribute to resistance to antitumor therapies, either by physical barriers (Vasan; Baselga; Hyman, et al., 2019) and/or by stimulating pro-tumor cells (Pitt et al., 2016).

Therefore, immune cells are key elements that can have anti-tumor or pro-tumor functions. Determining which function of immune cells will be predominant in the tumor microenvironment is associated with the phases of the immunoediting process. Immunoediting consists of three phases: elimination, equilibrium and escape (Teng; Kershaw; Smyth, 2013; Desai; Coxon; Dunn, 2022). During elimination, the action of anti-tumor immune cells predominates, in equilibrium, the immune cells of the adaptive immune response prevail, which promote tumor dormancy, and during escape, pro-tumor cells stand out (Lasek, 2022). This escape can be mediated by various factors, such as the non-recognition of anti-tumor immune cells by the class I/II human histocompatibility complex (MHC) (Keshavarz-Fathi; Rezaei, 2018).

Given this, immunotherapy aims to strengthen the immune system through active immunization and passive immunization (Abbott; Ustoyev, 2019). The type of immunization will depend on whether the individual has a residual immune response or the response is deficient or unresponsive, and it is indicated to strengthen the immune system by active immunization and passive immunization, respectively (Escribese; Barber, 2017). Among the types of active immunization, peptide vaccines stand out due to their cost-effectiveness, satisfactory immune tolerance, desired immunogenicity and easier monitoring of immune responses (Farran et al., 2019; Parvizpour et al., 2020).

Vaccines can be developed using the reverse vaccinology approach, which is based on genomic information to determine the most promising antigenic proteins that will be part of their constitution (Goodswen; Kennedy; Ellis, 2023). This screening is carried out using immunoinformatics, which is a series of tools that help to select the peptides most likely to trigger an immune response in populations that express a variety of alleles (Backert; Kohlbacher, 2015). In the case of vaccines developed to treat cancer, the presence of multiple epitopes increases the likelihood that the action will be more effective, as different epitopes can stimulate different immune cells.

Another important point in the development of a multi-epitope anti-tumor vaccine is to identify a target that is as specific as possible to mitigate side effects. From this conception, cancer testis antigens (CTAs) are viable targets, as they are only expressed in germ cells, trophoblasts and tumor cells (Nin; Deng, 2023). Silencing these antigens in normal cells allows the action of the multi-epitope vaccine to be directed at tumor cells. Several studies that have been developed using this approach with one antigen, multiple antigens or in combination with chemotherapy (Meng et al., 2021).

In addition to being expressed in tumor cells, CTAs are involved in several characteristics that favor the process of carcinogenesis, such as TFDP3, which acts in the activation of cell cycle and cell proliferation, differentiation and apoptosis of the cells in which they are present (Huang et al., 2021). The expression of the DP3 family transcription factor (TFDP3) has already been identified in triple negative breast cancer (Ding et al., 2018), prostate cancer (Ma et al., 2014), childhood T-cell lymphoblastic leukemia (Chun et al., 2017) and hepatocellular carcinoma (Wang et al., 2021). Therefore, TFDP3 was considered in this study for the development of a multi-epitope vaccine and interaction with Toll-like receptors (TLRs).

Sequence retrieval

The amino acid sequence of TFDP3 (UniProt ID: Q5H9I0) was obtained in FASTA format from UniProt (https://www.uniprot.org/) (accessed on 23 June 2023) database.

Prediction of linear B-cell epitopes

The recognition of B lymphocyte epitopes is a crucial step in the development of multi-epitope vaccines due antibodies released by these cells stimulate humoral immunity. The linear epitopes of B lymphocytes were predicted by ABCpred server (https://webs.iiitd.edu.in/raghava/abcpred/index.html) (Saha and Raghava, 2006), which uses the recurrent neural network approach and is about 65.93% accurate. The ABCpred server used a threshold of 0.51 and an amino acid sequence length of 16 for sequence evaluation.

Prediction of helper T-cell epitopes

T lymphocyte epitopes must interact with MHC class I and class II molecules for a better cellular immunogenic reactivity. The epitopes that bind to MHC class I and MHC class II molecules were predicted by the artificial neural network-based servers, NetMHCpan-4.1 (https://services.healthtech.dtu.dk/service.php?NetMHCpan-4.1) (accessed on 19 July 2023) and NetMHCIIpan-4.0 (https://services.healthtech.dtu.dk/service.php?NetMHCIIpan-4.0) (accessed on 19 July 2023) were used with default parameters (Reynisson et al., 2020). For NetMHCpan-4.1 server, peptides consisting of up to 9 amino acids were considered, and the % rank limits were set for strong binding < 0.50% and for weak binding < 2.00%. On the other hand, the NetMHCIIpan-4.0 server considered peptides made up of 15 amino acids, the % rank limits established for strong binding < 2% and for weak binding < 10%.

Prediction of antigenicity, allergenicity, and toxicity of epitopes

Epitopes identified for B lymphocytes and T lymphocytes were evaluated for antigenicity using VaxiJen v 2.0 server (http://www.ddg-pharmfac.net/vaxijen/VaxiJen/VaxiJen.html) (Doytchinova; Flower, 2007), with a threshold of 0.5 and selected a target "tumor". Allergenicity was acessed by AllerTop v2.0 server (https://www.ddg-pharmfac.net/AllerTOP/index.html) (Dimitrov et al., 2014), and toxicity by ToxinPred (https://webs.iiitd.edu.in/raghava/toxinpred/index.html) (Gupta et al., 2013), which used the "SVM (Swiss-Prot)" method and no threshold. Only linear B lymphocyte epitopes and T lymphocyte epitopes identified as antigenic, non-allergenic, and non-toxic were considered to make up the multi-epitope vaccine.

Prediction of interferon-gamma-inducing epitopes

The identification of epitopes that have an affinity for MHC class II and activate CD4 T lymphocytes through the stimulation of IFN-γ plays an important role in immune recognition and the development of the tumor immune response (Haabeth, 2014). IFNepitope server (https://webs.iiitd.edu.in/raghava/ifnepitope/predict.php) (accessed July 24, 2023) was used to identify epitopes with IFN-γ induction potential using the support vector machine (SVM) algorithm (Dhanda; Vir; Raghava, 2013).

Estimation of population coverage

The use of multi-epitopes in vaccines can increase population coverage due to the variety of alleles in the world's population. For this reason, the MHC class I and class II epitopes alleles were evaluated in covered population using the IEDB tool (http://tools.iedb.org/population/) (accessed on July 20, 2023) (Bui et al., 2006). The regions considered for evaluation were Europe, the United States, China, Japan, South America, and Brazil.

Design of the multi-epitope vaccine

The vaccine was developed with epitopes from B lymphocytes, MHC class I and class II. The epitopes selected are immunogenic, non-allergenic, non-toxic, and IFN-γ inducing (MHC class II). In the case of MHC class I and class II epitopes, those with the highest number of HLA alleles for each epitope and broad population coverage were also considered. In addition, epitopes that did not overlap in terms of amino acid position were screened to exploit the greater diversity of epitopes in TFDP3 sequence. The linkers EAAAK, KK, GPGPG, AAY, and a Histidine tail (6xHist) were used to join the epitopes (Nezafat et al., 2014). To increase immunogenicity, the 50S ribosomal protein L7/L12 from Mycobacterium tuberculosis Rv0652 was used as an adjuvant (Lee et al., 2014; Mahdevar et al., 2021).

Homology analysis of the multi-epitope vaccine with the host

Homology with proteins expressed in normal tissue other than tumor tissue can induce autoimmunity. The NCBI BLASTp online server (https://blast.ncbi.nlm.nih.gov/Blast.cgi) (Johnson et al., 2008) was used to check the homology of the multi-epitope vaccine with other human proteins. The vaccine amino acid sequence was added in FASTA format, the selected target organism was Homo sapiens (taxid:9606) and BLAST analysis was run, with the remaining parameters as default.

Evaluation of the physicochemical properties of the multi-epitope vaccine

Once the epitopes have been combined to form a vaccine, it is important to evaluate their characteristics, such as antigenicity, allergenicity, and toxicity. These factors contribute to the development of a safe immune response. The servers used for parameters evaluation were the same as used in B lymphocytes and T lymphocytes epitopes predictions (acesso em 20 de julho de 2023). In addition, the physical and chemical properties were evaluated by the Expasy ProtParam server (https://web.expasy.org/protparam/) (accessed on 20 October 2023) which included aliphatic index (indicates thermostability of the protein), hydropathicity index (GRAVY), molecular weight, isoelectric point (IP) and amino acid composition in vaccine (Gasteiger et al., 2005). The vaccine solubility was assessed by SOLPro (http://scratch.proteomics.ics.uci.edu/), which has 74.15% accuracy in support vector machine (SVM) prediction (threshold ≥ 0.5) (Magnan; Randall; Baldi, 2009).

Prediction of proteasome processing, interaction with TAP transporters (MHC class I biosynthesis), and post-translational modifications of the multi-epitope vaccine

The identification of cleavage sites via vaccine proteasome was predicted by NetChop-3.1 prediction method ("C term 3.0") with a threshold of 0.5 (https://services.healthtech.dtu.dk/services/NetChop-3.1/) (Kesmir et al., 2002). The binding affinity between the vaccine and the TAP transporters was also evaluated using TAPPred (https://webs.iiitd.edu.in/raghava/tappred/) (Bhasin; Raghava, 2004). In addition, post-translational modifications by glycosylation and phosphorylation were predicted by NetOGlyc v.4.0 (https://services.healthtech.dtu.dk/services/NetOGlyc-4.0/) (Steentoft et al., 2013) and NetPhos v. 3.1 (https://services.healthtech.dtu.dk/services/NetPhos-3.1/) (Blom; Gammeltoft; Brunak, 1999), respectively.

Secondary and tertiary structure prediction, refinement, and validation

The secondary structure influences the conformation in three-dimensional structure of the multi-epitope vaccine. Thus, the secondary structures, such as the beta-sheet and alpha-helix, of the vaccine's amino acid sequence were predicted by PSI-blast (PSIPRED 4.0) (http://bioinf.cs.ucl.ac.uk/psipred/), which uses two hidden layers of neural networks with 84.2% prediction accuracy (Jones, 1999; McGuffin; Bryson; Jones, 2000).

The tertiary structure of the multi-epitope vaccine was modeled using the ROBETTA server (RoseTTAFold method) (https://robetta.bakerlab.org/) (accessed September 14, 2023) (Baek et al., 2021). The quality of the models was assessed by ERRAT (https://saves.mbi.ucla.edu/), which analyzes the interaction between the atoms in structure by comparing them with X-ray crystallographic structures (Colovos, C.; Yeates, T. O., 1993). The model with the highest quality factor was refined using the GalaxyRefine server (https://galaxy.seoklab.org/cgi-bin/submit.cgi?type=REFINE), which uses loop modeling and molecular dynamics relaxation (Lee; Heo; Seok, 2015).

The model's validation was assessed using the Ramachandran plot obtained by PROCHEK (https://saves.mbi.ucla.edu/), which indicates residues in favorable and unfavorable regions (Laskowski; MacArthur; Thornton, 2012). In this case, the model has good quality if at least 90% of the residues are in favorable regions. The compatibility of the three-dimensional model with amino acid sequence itself was assessed by VERIFY 3D (https://saves.mbi.ucla.edu/) in which at least 80% of the amino acids must have a score >= 0.1 in 3D/1D profile to be considered a compatible model (Eisenberg; Lüthy, Bowie, 1997). ProSA-web (https://prosa.services.came.sbg.ac.at/prosa.php) was used to calculate the energy potential of the model, which indicates how close it is to a native protein (Wiederstein; Sippl, 2007). The best tertiary structure of the model was visualized using PyMol (Schrödinger; DeLano, 2020).

Analysis of discontinuous B-lymphocyte epitopes in the three-dimensional structure of a multi-epitope vaccine

Discontinuous B-lymphocyte epitopes can be identified in the three-dimensional structure of a protein. The prediction of these epitopes was obtained by the Ellipro server (http://tools.iedb.org/ellipro/) using the standard parameters (minimum score: 0.5 and maximum distance of 6 Ångströms (Ponomarenko et al., 2008). This server assigns a protrusion index (PI) based on the geometric properties of the protein's three-dimensional structure. The PI score determines the solvent accessibility of the residues, where higher is PI score, the greater the solvent accessibility of the residues.

Molecular docking analysis

Molecular docking between the multi-epitope vaccine and the immune system receptor plays an important role in predicting the interactions between them (Patra et al., 2020). Using this principle, the multi-epitope vaccine (ligand) was subjected to docking with Toll-like receptors (TLR), TLR2 (ID. PDB: 2z7x), TLR3 (ID. PDB: 2a0z), TLR4 (PDB: 3fxi), TLR7 (ID. PDB:7cyn) and TLR9 (ID. PDB: 8ar3) by ClusPro 2.0 (https://cluspro.bu.edu/) (Accessed: November 11, 2023), an automated algorithm that performs hard docking between proteins. The method presents the docked poses with strong complementarity and classifies them according to their clustering quality (Comeau et al., 2004). Subsequently, the pose with the lowest RSMD value for each receptor was visualized by PyMol and analyzed by the PDBSum Generation server (https://www.ebi.ac.uk/thornton-srv/databases/pdbsum/Generate.html) (Laskowski, 2009) to characterize the molecular interactions between the multi-epitope vaccine and the TLRs.

Molecular dynamics

Using molecular dynamics simulations, it is possible to predict, based on Newtonian physics, how the multi-epitope vaccine interacts with biological systems in aqueous media and indicating possible structural adaptations under physiological conditions (Aghajani, 2022). The molecular dynamics study of the multi-epitope vaccine interacting with Toll-like receptors (TLR2 and TLR3) used the GROMACS simulation package (Berendsen; Spoel; Drunen, 1995) available on the Visual Dynamics server (https://visualdynamics.fiocruz.br/en-US) (Vieira et al., 2023) with the Amber99 force field (Pearlman et al., 1995). The system was solvated from the TIP3P water model (Jorgensen et al., 1983) in a cubic box with boundaries 3 Ångströms from the edge of the protein. The system was neutralized with a 0.15 M NaCl solution to neutralize the negative charges of the amino acids. Energy minimization was carried out using the Steepest Descent algorithm (until a force of 1000 kJ.mol-1.nm-1 was reached) and Conjugate Gradient. The length of the van der Waals interactions and hydrogen bonds were fixed using the LINCS algorithm (Hess et al., 1997), allowing an integration time of 2fs to be used. The simulation was balanced using NVT for 20ps and NPT plus 20ps with constant temperature and pressure of 310K (optimum physiological temperature corresponding to 36.85°C) and 1 bar, respectively, using the Parrinelo-Rahman algorithm (Parrinelo; Rahman, 1981). The simulation time was 100ns with simulation steps of 2fs. From the molecular dynamics, data was obtained on the root mean square deviation (RSMD), root mean square fluctuation (RMSF), radius of rotation (Rg), and solvent accessible surface area (SASA) of the multi-epitope vaccine interacting with the TLRs throughout the simulation. Graphs were plotted using Python's numpy, seaborn, and matplotlib.pyplot packages to assess the stability of the multi-epitope vaccine throughout the simulation.

Codon optimization and in silico cloning

Codon optimization was carried out by converting the amino acid sequence to the nucleotide sequence using the JAVA Codon Adaptation server (https://www.jcat.de/) (Grote et al., 2005) with the Escherichia coli expression site for the multi-epitope vaccine. The metrics that measure the degree of translation of a protein, the codon adaptation index (CAI), and the CG component, were also obtained from this server. The ideal values for obtaining efficient protein expression are 1.0 and 30-70%, respectively for CAI and CG. Using the SnapGene software (https://www.snapgene.com/)(Accessed November 27, 2023), the sequence obtained was inserted into the 2707bp linearized Allele TA vector (Allele Biotechnology). This type of vector is used when convenient sites for the action of restriction enzymes are not located (Clark; Pazdernik; McGehee, 2019). In addition, the SnapGene software made it possible to obtain 1% agarose gels in TBE (Tris/Borate/EDTA) buffer solution and SB (sodium boric acid) buffer solution by simulating electrophoresis used to distinguish the insert, the vector, and the cloned vector.

Simulating the immune response

To predict the response of the immune system after administration of the multi-epitope vaccine, the C-ImmSim immune simulation server (https://kraken.iac.rm.cnr.it/C-IMMSIM/) (Accessed October 27, 2023) (Rapin et al., 2010) was used, which uses a position-specific scoring matrix and machine learning techniques to predict immune responses. This model simulates the three main functional components of the human immune system (bone marrow, thymus, and lymph node). This study considered one and three administrations of the multi-epitope vaccine containing 1000 antigens each at an interval of 4 weeks (recommended vaccination interval). Time steps were set at 1, 84, and 168 (each time step corresponds to 8 hours and a time step of 1 is equal to zero injection time). The alleles chosen correspond to those related to the epitopes of the multi-epitope vaccine: MHC class I (A0301, A1101, B1501, and B3501) and MHC class II (DRB1_0401 and DRB1_0402). A value of 1050 was adopted for the "Simulation Steps" parameter and the others followed the standards already established.

Prediction and selection of epitopes

In this study, 405 amino acid sequence of TFDP3 (Uniprot ID: Q5H9I0), epitopes were predicted for B cells, CD8+ T lymphocytes and CD4+ T lymphocytes. The identification of B lymphocyte epitopes will induce the release of antibodies, stimulating humoral immunity. The epitopes of CD8+ T lymphocytes and CD4+ T lymphocytes interact with MHC class I and class II molecules, respectively, and can trigger the activation of dendritic cells, production of IFN-γ and induction of the apoptosis process in malignant neoplastic cells (Jahangirian et al., 2022). In addition, the epitopes of these cell types must have antigenicity, non-allergenicity, non-toxicity and the ability to induce the production of interferon-gamma, specifically the latter in the case of CD4+ T lymphocytes.

A total of 40 linear epitopes (each 16 amino acids long) of antigenic B lymphocytes were identified, half of which showed probable antigenicity (Table S1). Of these epitopes, 14 were classified as probably non-allergenic and non-toxic (Table 1).

Table 1: Predicted B cell epitopes and their immunogenic properties.

Sequence	Position	Vaxijen score	Allergenicity	Toxicity
TSSGGSQYSGSRVETP	368-383	1.0805	Non-allergenic	Non-toxic
RVERQKRLERIKQKQS	201-216	1.6392	Non-allergenic	Non-toxic
CSISDDKSEYLFKFNS	269-284	0.7976	Non-allergenic	Non-toxic
SSPPWAGQHNRKGEKN	92-107	0.7683	Non-allergenic	Non-toxic
DVKNIKRRTYDALNVL	156-171	1.0012	Non-allergenic	Non-toxic
SGSCSAEDLKMARNLV	306-321	0.7467	Non-allergenic	Non-toxic
PKALEPYVTEMAQGTF	322-337	0.5497	Non-allergenic	Non-toxic
LMDENQTSRPVAVHTS	17-32	0.5233	Non-allergenic	Non-toxic
LSMKVWETVQRKGTT	114-129	0.7633	Non-allergenic	Non-toxic
EDLKMARNLVPKALEP	312-327	0.5484	Non-allergenic	Non-toxic
PQRPAASNIPVVGSPN	62-77	0.8037	Non-allergenic	Non-toxic
TNSAQNCQNLRVERQK	191-206	0.9420	Non-allergenic	Non-toxic
PNESAYDVKNIKRRTY	150-165	0.6101	Non-allergenic	Non-toxic
FRAASNHASPNESAYD	141-156	0.5059	Non-allergenic	Non-toxic

Regarding the epitopes of CD8+ T lymphocytes for MHC class I, 17 linear epitopes (length of 9 amino acids each) were identified that showed probable antigenicity in relation to the representative HLA-A and HLA-B (Human Leukocyte Antigens) alleles (Table S2). Of these epitopes, 8 were classified as probable non-allergenic and probable non-toxic (Table 2).

Table 2: Predicted epitopes of CD8+ T lymphocytes for MHC class I antigens their immunogenic properties.

Sequence	Position	MHC I Alleles	Vaxijen score	Allergenicity	Toxicity
KVWETVQRK	117-125	HLA-A03:01 HLA-A11:01 HLA-A30:01 HLA-A31:01	0.7471	Non-allergenic	Non-toxic
DALNVLMAM	166-174	HLA-A26:01 HLA-B35:01	0.8599	Non-allergenic	Non-toxic
KRLERIKQK	206-214	HLA-B*27:05	1.5955	Non-allergenic	Non-toxic
RIKQKQSEL	210-218	HLA-B07:02 HLA-B08:01 HLA-A*30:01	1.5326	Non-allergenic	Non-toxic
SISDDKSEY	270-278	HLA-A01:01 HLA-A26:01 HLA-B15:01 HLA-B35:01	0.6114	Non-allergenic	Non-toxic
ISDDKSEYL	271-279	HLA-A*01:01	0.9410	Non-allergenic	Non-toxic
AGSRSNGTW	344-352	HLA-B*58:01	0.5672	Non-allergenic	Non-toxic
ATSSGGSQY	367-375	HLA-A01:01 HLA-A11:01 HLA-A26:01 HLA-B15:01	1.3318	Non-allergenic	Non-toxic

HLA - Human Leukocyte Antigen, *- indicates that HLA typing and its corresponding location in chromosome.

When the epitopes of CD4+ T lymphocytes for MHC class II were evaluated, 6 linear epitopes (length of 15 amino acids) were identified that showed probable antigenicity in relation to the representative HLA-DR alleles (Table S3). Of these epitopes, four were classified as probably non-allergenic, non-toxic, and INF-γ inducing (Table 3).

Table 3: Predicted epitopes of CD4+ T lymphocytes for MHC class II antigens their immunogenic properties.

Sequence	Position	MHC II Alleles (HLA)	Vaxijen score	Allergenicity	Toxicity	INF- γ Inducers
ELKVLMDENQTSRPV	91-105	DRB103:01DRB103:05DRB104:01DRB104:02DRB114:02DRB301:01	0.6150	Non-allergenic	Non-toxic	Inducer
SPPWAGQHNRKGEKN	171-185	DRB101:03 DRB104:01 DRB104:02 DRB104:08	0.8473	Non-allergenic	Non-toxic	Inducer
VGELVAKFRAASNHA	212-226	DRB111:01 DRB111:04 DRB1*13:01	0.6633	Non-allergenic	Non-toxic	Inducer
VAKFRAASNHASPNE	216-230	DRB101:01 DRB101:02 DRB101:03 DRB104:01 DRB104:02 DRB104:03 DRB104:04 DRB104:05 DRB104:08 DRB107:01 DRB108:01 DRB108:03 DRB109:01 DRB110:01 DRB111:01 DRB113:02 DRB113:03 DRB114:02 DRB316:01 DRB302:02 DRB5*01:01	0.6149	Non-allergenic	Non-toxic	Inducer

HLA - Human Leukocyte Antigen.

Analysis of the population distribution of the predicted epitopes for MHC class I and class II

The high polymorphism of MHC class I and class II proteins play a key role in adaptive immunity (Wieczorek et al., 2017). HLA alleles have a different distribution in different countries and geographic regions of the world. Because of these factors, the analysis of the population coverage of epitopes is an important step for their validation as eligible to compose a multi-epitope vaccine. The epitopes evaluated were those that did not overlap in relation to the position of the amino acids to explore more distinct regions of the amino acid sequence of TFDP3. In relation to MHC class I, the following epitopes were evaluated: KVWETVQRK (117-125), ATSSGGSQY (367-375) and SISDDKSEY (270-278). In MHC class II, the epitopes were: ELKVLMDENQTSRPV (91-105), SPPWAGQHNRKGEKN (171-185) and VAKFRAASNHASPNE (216-230). The combination of epitopes using IEDB tool showed a population coverage of 93.55% worldwide, 95.88% in Europe, 94.93% in the United States, 85.48% in China, 88.94% in Japan, and 79.29% in South America (Supplementary Figure 1; Figure 1). In Brazil, population coverage is 84.45% (Figure 01). Considering the population coverage in relation to ethnicity in Brazil, it was 67.87% for Amerindians (indigenous people of the American continent), 95.06% for Caucasoids, 93.85% for mixed race and 79.84% for mulatto (Supplementary Figure 2).

Construction of the multi-epitope vaccine

The multi-epitope vaccine was designed using epitopes that did not overlap, had broad population coverage, and immunogenic characteristics. The selected epitopes of B lymphocytes were: RVERQKRLERIKQKQS (201–216), TNSAQNCQNLRVERQK (191–206), SGSCSAEDLKMARNLV (306–321). On the other hand, the MHC epitopes of class I and class II were those evaluated in population coverage: MHC class I: KVWETVQRK (117-125), ATSSGGSQY (367-375) and SISDDKSEY (270-278); MHC class II: ELKVLMDENQTSRPV (91-105), SPPWAGQHNRKGEKN (171-185) and VAKFRAASNHASPNE (216-230). These epitopes have been joined by ligands widely used in the construction of peptide vaccines that can make the protein stable (Nezafat et al., 2014), increase immunogenicity (Avvagari et al., 2020), and aid in vaccine purification (Herrera, 2020). The ligands used were EAAAK in the N-terminal portion binding to the adjuvant, KK among the epitopes of B lymphocytes, GPGPG among MHC class II epitopes (CD4+ T lymphocyte), AAY among MHC class I epitopes (CD8+ T lymphocyte) and a Histidine tail (6xHist) in the terminal C portion. The adjuvant used was Mycobacterium tuberculosis Rv0652 ribosomal protein L7/L12 (MAKLSTDELLDAFKEMTLLELSDFVKKFEETFEVTAAAPVAVAAAGAAPAGAAVEAAEEQSEFDVILEAAGDKKIGVIKVVREIVSGLGLKEAKDLVDGAPKPLLEKVAKEAADEAKAKLEAAGATVTVK) with the aim of increasing immunogenicity (Lee et al., 2014; Mahdevar et al., 2022) (Figure 2). After joining the epitopes, a vaccine with a sequence of 291 amino acids was obtained.

Homology and physicochemical properties of the vaccine

The homology of the multi-epitope vaccine in relation to human proteins showed identity of 35.86% with TFPD3, 49.32% with E2F protein and 40.62% with transcription factor Dp-1 and its isoforms (Table 4). The alignment of the proteins that presented identity with the multi-epitope vaccine presents a percentage ranging from 26% to 13% of coverage. Being 26% corresponds to the target of the study, that is, it has a low risk of developing autoimmunity and affecting normal tissues.

Table 4: Homology of the vaccine in relation to human proteins.

Description	Max Score	Total Score	Query Cover	E value	Per. ident	Acc. Len
Transcription factor of Dp family member 3 (NP_057605.3)	67.8	107	26%	2E-11	35.86%	405
Proteína E2F-like (AAF37562.2)	65.5	104	22%	8E-11	49.32%	345
Transcription factor Dp-1 isoform X6 (XP_016876208.1)	40.0	40.0	13%	0.019	40.62%	337
Transcription factor Dp-1, isoform CRA_b (KAI2569993.1)	39.7	39.7	13%	0.016	40.62%	236
Transcription factor Dp-1(EAX09217.1)	39.7	39.7	13%	0.022	40.62%	312
Transcription factor Dp-1(NP_009042.1)	39.7	39.7	13%	0.024	40.62%	410
Transcription factor Dp-1 isoform X4 (XP_005268385.1)	39.7	39.7	13%	0.024	40.62%	406
Transcription factor Dp-1(AAN46090.1)	39.7	39.7	13%	0.026	40.62%	406
Transcription factor Dp-1 isoform X2 (XP_047286517.1)	39.3	39.3	13%	0.035	40.62%	472
Transcription factor Dp-1 isoform X1 (XP_016876206.1)	39.3	39.3	13%	0.035	40.62%	476

The multi-epitope vaccine is antigenic (Vaxijen v.2 = 0.7018), non-allergenic, and has a low risk of toxicity (only the KKTNSAQNCQ peptide sequence (position: 154) (SVM score = 0.04) showed low toxicity). Regarding the physicochemical properties, the multi-epitope vaccine has a molecular weight of 31,142.24 Da, an isoelectric point of 9.05 that can be used in purification through isoelectric precipitation. The solubility of the vaccine after overexpression in Escherichia coli was about 0.9609 (96.1%). In addition, it has an instability index of 31.94 that classifies the vaccine as stable, an aliphatic index of 71.24 indicating thermal stability, and a hydropathy index (GRAVY) of –0.619 that classifies it as hydrophilic. The estimated half-life was 1 hour in mammalian reticulocytes (in vivo), 30 minutes in yeast (in vivo) and greater than 10 hours in E. coli (in vivo).

Processing via proteosome, interaction with TAP transporters (MHC class I biosynthesis) and post-translational modifications of the vaccine

Following the multi-epitope vaccine, 85 cleavage sites were predicted at the binding site of C-terminal regions by the proteosome (Table S4). On the other hand, in the interaction with the TAP transporters, 96 peptides with high binding affinity were predicted when considering the sequence and properties of the amino acids of the multi-epitope vaccine (Table S5). When the post-translational modifications of the multi-epitope vaccine were evaluated, 7 O-glycosylation sites (position 153 with score of 0.67, position 156 with score of 0.5, position 205 with score for 0.62, position 206 with score for 0.64, position 215 with score of 0.90, position 242 with score of 0.51 and position 279 with score of 0.61) and 26 phosphorylation sites in which 15 of them can be targeted by more than one kinase (Table S6; Figure 3).

Secondary and tertiary structure, refinement and validation

The prediction of the secondary structure of the amino acid sequence of the multi-epitope vaccine by PSIPRED server indicated the presence of 7.56% of beta-leaf, 39.52% of regions without defined secondary structures, and 52.92% of alpha-helix (Figure 4). These results indicate a predominance of secondary alpha-helix structure.

On the other hand, the tertiary structure was determined using Robetta server with RoseTTAFOLD method, whose five models showed a confidence of 0.55 (Supplementary Figure 3). From then on, these models were evaluated for stereochemical quality by ERRAT, PROCHECK and ProSA-Web tools. According to the validation by ERRAT, model 4 presented the highest quality factor corresponding to 99.293%, showing that the model has a high resolution (Supplementary Figure 4; Figure 5A). To improve the quality of model 4, GalaxyRefine was used, which presented five models (Table S7), in which model 3, when evaluated by ERRAT presented a quality factor of 99.647. This value corresponds to a percentage increase in model quality of 0.354 compared to model 4 predicted by the Robetta server (Figure 5B). When model 3 was compared to model 4 before refinement, there was an increase in Clash score from 1.7 to 13.8, an increase in a favorable Rama value from 95.2 to 96.5, an increase in the MolProbity from 1.259 to 1.873, the RSMD went from 0 to 0.24, the GDT-HA went from 1 to 0.9923, and there was no bad rotamer (Table S7).

Then, model 3 obtained by refinement was evaluated by PROCHECK, VERIFY-3D and ProSA-web. The Ramachandran graph was obtained using PROCHECK, which showed 93.7% of the waste in favorable regions, 4.7% of the waste in additional permitted regions, and 0.8% of the waste in non-permitted regions (Figure 6 A). The three-dimensional structure was compatible with the amino acid sequence, since 80.76% of the residuals had a mean score of 3D-1D >=0.1 by Verify 3D server (Figure 6 B). In the graph obtained by ProSA, the z-score obtained from the structure was -6.98, which indicates that the structure is closer to the native structure of the protein (Figure 6, C-D). Figure 6 E shows the tertiary structure of model 3 obtained by refinement.

Identification of discontinuous B cell epitopes in the three-dimensional structure of multi-epitope vaccine

Identification of the regions of interaction between B lymphocyte epitopes and antibodies plays a key role for development of an immune response (Kringelum et al., 2012). In the three-dimensional structure of the multi-epitope vaccine, 6 linear epitopes were predicted, with size ranging from 41 to 4 residues and a score (PI) from 0.819 to 0.542 (Table 05). Regarding discontinuous epitopes, 6 epitopes were also predicted, with a size ranging from 62 to 6 residues and a score (PI) from 0.822 to 0.513 (Table 5). There was an overlap in part of residues of the linear epitopes with the residues of discontinuous epitopes, as in epitope 1 (204-209) and epitope 7 (207-233) (Figure 7), with high score values (PI), which indicates a presence of epitopes that make the multi-epitope vaccine more immunogenic.

Table 05: Linear epitopes of B lymphocytes of the three-dimensional structure of the multi-epitope vaccine.

Epitopes	Position	Peptides	Number of residues	Score (PI)
1	204-249	QTSRPVGPGPGSPPWAGQHNRKGEKN	26	0.819
2	70-110	VILEAAGDKKIGVIKVVREIVSGLGLKEAKDLVDGAPKPLL	41	0.802
3	116-135	EAADEAKAKLEAAGATVTVK	20	0.708
4	266-291	ISDDKSEYAAYATSSGGSQYHHHHHH	26	0.688
5	164-183	NLRVERQKKKSGSCSAEDLK	20	0.63
6	1-4	EAAA	4	0.542

PI – protrusion index.

Table 06: Discontinuous epitopes of B lymphocytes of the three-dimensional structure of the multi-epitope vaccine.

Epitopes	Residues	Number of residues	Score (PI)
7	A:R207, A:P208, A:V209, A:G210, A:P211, A:G212, A:P213, A:G214, A:S215, A:P216, A:P217, A:W218, A:A219, A:G220, A:Q221, A:H222, A:N223, A:R224, A:G226, A:E227, A:K228, A:N229, A:P233	23	0.822
8	A:V70, A:I71, A:L72, A:E73, A:A74, A:A75, A:G76, A:D77, A:K78, A:K79, A:I80, A:G81, A:V82, A:I83, A:K84, A:V85, A:V86, A:R87, A:E88, A:I89, A:V90, A:S91, A:G92, A:L93, A:G94, A:L95, A:K96, A:E97, A:A98, A:K99, A:D100, A:L101, A:V102, A:D103, A:G104, A:A105, A:P106, A:K107, A:P108, A:L109, A:L110, A:E111, A:A114, A:E116, A:A118, A:D119, A:E120, A:A121, A:K122, A:A123, A:K124, A:L125, A:E126, A:A127, A:A128, A:G129, A:A130, A:T131, A:V132, A:T133, A:V134, A:K135	62	0.762
9	A:N161, A:N164, A:L165, A:V167, A:E168, A:R169, A:Q170, A:K171, A:K172, A:K173, A:S174, A:G175, A:S176, A:I266, A:S267, A:D268, A:D269, A:K270, A:S271, A:E272, A:Y273, A:A274, A:A275, A:A277, A:T278, A:S280, A:G281, A:G282, A:S283, A:Q284, A:Y285, A:H286, A:H287, A:H288, A:H289, A:H291	36	0.688
10	A:C177, A:S178, A:A179, A:E180, A:D181, A:K183, A:A245, A:S246, A:N248, A:E249	10	0.618
11	A:K197, A:M200, A:D201, A:Q204, A:T205, A:S206, A:A240, A:N243, A:H244	9	0.613
12	A:E1, A:A2, A:A3, A:A4, A:K8, A:L9	6	0.513

PI – protrusion index

Molecular anchoring of multi-epitope vaccine with TLR-2, TLR-3, TLR-4, TLR-7 and TLR-9

Toll-like receptors (TLRs) are molecules that mediate the response in innate immunity in diverse disease conditions. There is evidence that these molecules are expressed in malignant neoplastic cells with a vital role in tumor progression (Kumar et al., 2022). The interaction of the multi-epitope vaccine with TLRs may enhance immune response. Therefore, the molecular docking interaction of the multi-epitope vaccine with TRL-2, TLR-3, TLR-4, TLR-7 and TLR-9 molecules was evaluated using ClusPro server. The binding affinity energy of TLRs best pose: TLR-2 (–691.4 Kcal/mol), TLR-3 (–806.9 Kcal/mol), TLR-4 (-705.6 Kcal/mol), TLR-7 (-1485.4 Kcal/mol) and TLR-9 (-1002.0 Kcal/mol). The PyMol software was used to visualize the complex generated by docking (Figure 8).

The identification of the molecular interactions between the vaccine and the TLRs was achieved by the PDBSum Generate server which identified a total of 23 hydrogen bonds, 8 saline bridges, and 190 non-contact interactions for TLR-2 in the C-chain, and 2 hydrogen bonds, 1 saline bridge, and 20 non-contact interactions for the TLR-2 D-chain (Figure 9A). For TLR-3 in the B chain, there are 13 hydrogen bonds, 12 saline bonds, and 217 non-contact interactions (Figure 9B). The presence of salt bridges and non-contact interactions in the interaction of the vaccine with TLRs are essential for the maintenance of protein structure and stability.

In addition, molecular interactions between the vaccine and TLR-4, TLR-7 and TLR-9 were identified. In TLR-4, 5 hydrogen bonds, 1 salt bridge and 52 non-contact interactions in the F chain were identified, and the presence of 5 hydrogen bonds and 68 non-contact interactions in the H-chain (Figure 9C). In TLR-7, 2 hydrogen bonds and 7 non-contact interactions in the I-chain, 1 hydrogen bond and 32 non-contact interactions in the L-chain, and 5 hydrogen bonds, 6 saline bridges and 177 non-contact interactions in the M-chain were identified (Figure 9D). In TLR-9, 10 hydrogen bonds, 1 saline bridge and 129 non-contact interactions in the N chain were identified (Figure 9E).

Evaluation of the stability in multi-epitope vaccine by molecular dynamics

Molecular dynamics is fundamental to assess the stability of molecules in a biological system. During the molecular dynamics simulations, structural variations of the multi-epitope vaccine, when interacted with Toll-like receptors (TLR2 and TLR3), were evaluated using the metrics of RMSD (root square of mean square deviation), RMSF (square root of quadratic fluctuation), Rg (radius of rotation) and SASA (surface area accessible to solvent).

The RSMD is the value used to calculate deviation in the protein backbone (Cα, C, and N) during the 100 ns. When the multi-epitope vaccine interacted with the TLRs, it showed an increase in RSMD at the beginning of the simulation, when an increase in temperature occurred, and then stabilized, indicating the equilibrium of the system. The multi-epitope vaccine in the interaction with TLR2 presented an RMSD with a mean value of approximately 0.86 nm (approximate standard deviation = 0.20 nm), while with TLR3 the approximate mean value was 0.46 nm (approximate standard deviation = 0.09 nm). After 20 ns of simulation, the multi-epitope vaccine tended to be stable in relation to both TLR2 (approximate mean value = 0.83 nm and approximate standard deviation = 0.17 nm) and TLR3 (approximate mean value = 0.47 nm and approximate standard deviation = 0.07 nm) (Figure 10 A-B). These results indicate that the multi-epitope vaccine has a higher stability after 20 ns, especially when it interacted with TLR3, as it has the lowest mean RSMD value.

With RSMF of alpha carbon (Cα) atoms it is possible to determine which amino acids of the multi-epitope vaccine are highly flexible over the simulation time. The higher RMSF values are usually associated with loops, and the lower ones with helices. In the graph of multi-epitope vaccine interaction with TLR2, residues 96 (0.7146 nm), 213 (0.6714 nm), 214 (0.6672 nm), 218 (0.6206 nm) and 287 (0.6206 nm) showed the highest RSMF values, which correspond in most of these residues to regions where loops predominate, which explains the flexibility of the molecule in these regions (Figure 10C). On the other hand, in relation to the multi-epitope vaccine with TLR3, the residues 279 (0.7442 nm), 281 (0.7042 nm), 283 (0.6814 nm), 284 (0.7506 nm) and 291 (0.6907 nm) located closer to the C-terminal portion presented the highest RSMF values and correspond as in TLR2 to amino acids located in the most flexible regions of the multi-epitope vaccine (Figure 10D). In both the interaction with TLR2 and TLR3, the multi-epitope vaccine presents flexible regions during simulation.

On the other hand, the Radius of Gyration (Rg) is related to the degree of folding of the multi-epitope vaccine throughout the simulation. The approximate mean value of the total turning radius of the multi-epitope vaccine in relation to the interaction with TLR2 was 2.67 nm (approximate standard deviation of 0.05 nm) and from 20000 ps of 2.66 nm (approximate standard deviation of 0.05 nm) (Figure 10E). Regarding the interaction with TLR3, the approximate mean value was 3.28 nm (approximate standard deviation 0.05 nm) and from 20000 ps onwards 3.27 nm (approximate standard deviation 0.04 nm) (Figure 10F). These data reveal that the multi-epitope vaccine, when interacting with TLR2, presents a higher degree of folding during the simulation, since its total Rg value was lower than the value of the interaction with TLR3.

The Solvent Accessible Surface Area (SASA) is a parameter that determines the surface of the multi-epitope vaccine exposed to the solvent during the simulation. SASA in the multi-epitope vaccine in relation to interaction with TLR2 presented an approximate mean value of 170.02 nm² (approximate standard deviation 4.78 nm²) and with TLR3, 294.66 nm² (approximate standard deviation 3.18 nm²) (Figure 10G-H). These results indicate that the multi-epitope vaccine, when it interacts with TLR3, has a more stable behavior in the environment with the solvent, since it presents an approximate mean SASA value higher than that found for TLR2.

The parameters evaluated in molecular dynamics showed that the interactions of the multi-epitope vaccine with TLR3 are more stable than with TLR2 during the simulation, except for the total radius of gyration, which was larger in relation to TLR3, indicating a lower packing. It is likely that the nature of interactions established between the multi-epitope vaccine and TLR3 and the more flexible regions of the multi-epitope vaccine have an influence on this packaging.

Codon adaptation and in silico cloning of the multi-epitope vaccine

Adaptation of codons is usually necessary when there is an interest in increasing the efficiency of the target gene. With the JAVA Codon Adaptation server, the amino acid sequence of the multi-epitope vaccine converted into nucleotide sequence was optimized in relation to E. coli (K12 strain) and obtained with an approximate value of 50.73% of CG content contained in the optimal range of 30-70%. In addition, the codon adaptation index (CAI) was 1.0, which corresponds to an ideal value for efficient protein expression. Both parameters reinforce that the designed vaccine has high transcriptional and translational efficiency. After the optimization of the multi-epitope vaccine, cloning was performed in the linearized Allele TA vector of 2707bp using the SnapGene software. The final length of the cloned vector was 3581bp, which corresponds to 2707bp of the vector plus the insert (amplified from the multi-epitope vaccine) of 873bp (Figure 11A). In order to visualize the distinction of the insert, vector and cloned vector when observing the 1% agarose gel of the electrophoresis simulation (Figure 11B-C) it was verified that the size of the cloned vector is lower than what was expected when in TBE buffer, but in low ionic strength (SB) sodium boric acid buffer it presents high resolution.

Simulation of the immune response after administration of the multi-epitope vaccine

In the simulation of the immune response triggered by the single administration and three doses of the multi-epitope vaccine, there was a lasting and more robust immune response in repeated exposure to the antigen. Considering that innate immunity is the primary step of the immune response involved in the presentation of the antigen and activation of T-cells. The multi-epitope vaccine activated and stimulated natural killer (NK) cells and macrophages in both vaccination schedules (Figure 12A, B, C and D). An increase in the level of cytokines was also observed in both vaccination schedules (Figure 12E and F). In the three administrations, there was an increase in cytokine interferon gamma (IFN-γ) with a tendency to decrease in the following exposures and a gradual increase in IL-2 (Figure 12F).

In addition, after both vaccination schedules, there was a decrease in activated macrophages, which indicates the probable action of anti-inflammatory cytokines such as IL-10 and TGF-β to control the immune response and prevent the development of an exacerbated inflammatory response (Figure 12C, D, E and F).

Regarding T-cell populations, the multi-epitope vaccine increased the Th1 subpopulation in both vaccine schedules (Figure 13A, B, C and D). Successive exposure to antigens kept the amount of Th1 cells elevated for longer. Th1 lymphocytes are induced by cytotoxic T lymphocytes that proliferate and increase cytotoxic capacity. This event can be observed by the increased activation of the cytotoxic T lymphocyte population and a decrease in resting cells with the immune response induced by the multi-epitope vaccine in both vaccine schedules (Figure 13E, F, G and H). However, in three administrations the T lymphocyte population remained activated longer compared to one administration (Figure 13H).

In the immune response mediated by the B lymphocyte population, there was an increase in memory cells in both vaccine schedules, with differentiation and production of IgM and IgG, and with a decrease in naïve B lymphocytes (Figure 14A, B, C and D). In addition, a gradual increase in the production of IgM+IgG, IgM, IgG1 and IgG1+IgG2 was identified with the administration of the injections at the pre-established intervals of the multi-epitope vaccine (Figure 14E and F). During the successive exposure of the multi-epitope vaccine, there was a more robust immune response. These results demonstrate that the multi-epitope vaccine with successive vaccine exposure has the potential to cause an effective and long-term immune response.

During the development of therapeutic cancer vaccines, antigen selection is one of the most important study steps (Liu et al., 2022). In peptide-based vaccines, autologous antigens can be used, such as cancer-testis antigens (CTA), i.e., antigens expressed in malignant tumor cells, embryonic cells, testicular germ cells, and have reduced expression in other normal tissues (Lam et al., 2021).

Several immunoinformatics studies that have shown promising results from CTA-based multi-epitope vaccines, such as in ovarian cancer (Sufyan et al., 2021), breast cancer (Krishnamoorthy, H. R.; Karuppasamy, R., 2023) and in prostate cancer (Patra, et al., 2020). Regarding clinical trials, there are peptide vaccines already evaluated for melanoma (Hu et al., 2015), prostate cancer (Sonpavde et al., 2014) and ovarian cancer (Diefenbach et al., 2008) that have shown positive results.

The cancer-testis antigen TFDP3 is a transcription factor expressed in multiple cancers, such as in prostate cancer (Ma et al., 2014), childhood T-cell lymphoblastic leukemia (Chu et al., 2017), hepatocellular carcinoma (Wang et al., 2021), and breast cancer (Yin et al., 2017; Ding et al., 2018). In addition, this CTA is involved in important physiological processes that favors the development of cancer, such as cell proliferation (Huang, J. et al., 2021), regulation of mesequimal epithelial transition (Yin et al., 2017), modulation of cell apoptosis (Ding et al., 2018), and chemoresistance in residual disease (Chu et al., 2017).

Based on this, in this study, TFDP3 was considered for the prospection of epitopes with the ability to trigger humoral and cellular immune responses, and to build a multi-epitope therapeutic vaccine against cancer. Because of the high costs in the development of vaccines, the use of immunoinformatics tools was the best choice for the development of the multi-epitope vaccine. Immunoinformatics consists of numerous viable, accurate, and rapid in silico tools for the study and development of multi-epitope vaccines in various chronic diseases, including cancer (Bahrami et al., 2019).

Peptide-based vaccines developed for cancer treatment depend on the interaction of T lymphocytes and B lymphocytes, as they have antitumor action (Zhang et al., 2019). In this study, immunoinformatics analysis revealed several antigenic and immunogenic epitopes with the ability to stimulate the humoral (B lymphocyte) and cellular (CD8+ and CD4+ T lymphocyte) response and to induce the production of INF-γ by the epitopes of CD4+ T lymphocytes. INF-γ is responsible for the activation of the immune system and the antitumor response by several mechanisms, such as the induction of apoptosis regulated by T cells, inhibition of angiogenesis, and stimulation of pro-inflammatory M1 macrophages (Jorgovanovic et al., 2020).

CD8+ and CD4+ T lymphocytes are associated with MHC class I and class II, respectively, which are highly polymorphic in terms of their alleles (human leukocyte antigens (HLA)) (Rock; Reits; Neefjes, 2016). In tumor cells, tumor heterogeneity and several mechanisms that occur with tumors, such as rejection, escape, and dormancy, are associated with diversity or loss of expression of MHC class I/II alleles (Garrido; Aptsiauri, 2019). A variety of HLA alleles expressed for MHC class I and class II are expressed at different frequencies for different ethnicities (Bui et al., 2006).

Because of that, the epitopes associated with CD8+ and CD4+ T lymphocytes needed to be evaluated in terms of their population coverage, i.e., the prediction of which populations will benefit most from the multi-epitope vaccine consisting of these epitopes. The epitopes selected in this study to compose the multi-epitope vaccine showed a high population coverage worldwide (93.55%) and in other populations they range from 67.87% to 95.88%, emphasizing the multi-epitope vaccine as a potential immunotherapeutic agent for cancer types that express TFDP3.

The population coverage in the world identified in this study is in the range found in the prediction of multi-epitope vaccines for small cell lung cancer at 83.81% (Herrera, 2020), breast cancer of 90.33% (Krishnamoorthy; Ramanathan, 2023), ovarian cancer at 97.59% (Sufyan, et al., 2021), and melanoma between 93.55% to 99.13% (Safavi et al., 2019). These data reveal that the mutli-epitope vaccine in this study has a wide population coverage worldwide.

In this study, the multi-epitope vaccine was composed of 9 epitopes from the TFDP3 amino acid sequence that were more likely to trigger an immune response, mediated by B lymphocytes, CD4+ and CD8+ T lymphocytes, in several populations. In addition to these epitopes, it was necessary to insert the adjuvant and the ligands between the epitopes and at the N-terminal and C-terminal ends into the composition of the multi-epitope vaccine. The embedded Mycobacterium tuberculosis 50S ribosomal L7/L12 adjuvant can trigger an adaptive immune response (Rahmani et al., 2019). The ligands used in the multi-epitope vaccine aid in the maintenance of conformation-dependent immunogenicity, separation, and processing of epitopes for the antigen presentation process (Livingston et al., 2002; Mahdevar et al., 2022).

In addition, the multi-epitope vaccine had a low risk of inducing autoimmunity due to the low percentage of homology with other human proteins, antigenicity, non-allergenicity, and low risk of toxicity. It also has characteristics that can be useful in isolation by isoelectric precipitation and action in biological systems in vivo and in vitro, such as the isoelectric point of 9.05, good solubility in water, thermal stability and half-life of 1h in mammalian reticulocytes (in vivo).

The stability and solubility of recombined proteins are important factors in the production of vaccines, post-production, conservation of biological activity and reduction of toxic effects (Vazquez; Corchero; Villaverde, 2011). Based on these characteristics, including the hydrophilicity of the multi-epitope vaccine, the integrity of the vaccine in the face of the metabolic enzymes of the biological environment can be preserved until it is delivered to the cells using lipid nanoparticles as a vehicle for the application of the vaccine, as occurs in mRNA vaccines (Wilson; Geetha, 2022).

Regarding the processing considering MHC class I biosynthesis, the multi-epitope vaccine presented cleavage sites by proteosomes and peptides with high affinity for TAP transporters. In MHC class I biosynthesis, proteins are cleaved by the proteosome, can travel to the endoplasmic reticulum and interact with the TAP transporter that will assemble the peptide-class I complex, go to the Golgi complex, and be expressed on the cell surface (Rock.; Reits; Neefjes, 2016). Another important aspect evaluated in the multi-epitope vaccine was the identification of post-translational modifications by O-glycosylation and phosphorylation that can be used as indicators of peptide degradation in the proteosome (Zarling et al., 2000; Mahdevar et al., 2022).

From the point of view of secondary structure, the multi-epitope vaccine showed a predominance of alpha-helix conformation in relation to beta-leaf and regions without defined secondary structure. With tertiary structure modeling, it is possible to estimate protein dynamics, function, and ability to interact with other proteins (Sufyan et al., 2021). In the tertiary structure of the multi-epitope vaccine, the quality parameters were predicted and the structure was refined to obtain the best percentage values. By evaluation by the Ramachandran graph, Errat quality factor and the ProSA z-score, it was revealed that the multi-epitope vaccine presented a biologically compatible predicted structure. The tertiary structure of the multi-epitope vaccine also retained epitopes capable of interacting with B lymphocytes and developing a humoral response.

After the determination of the tertiary structure, the interaction of the vaccine with Toll-like receptors (TLRs) was identified, specifically TLR2 and TLR3 showed more interactions and, subsequently, the molecular dynamics were performed. TLR receptors are receptors present on the plasma membrane and membranes of immune and non-immune cells whose activation by some agent can lead to the production of cytokines, chemokines and growth factors that can help induce immune responses (McCall; Muccioli; Benencia, 2020).

TLR2 is commonly expressed in the cell membrane, but after cell transformation to dysplasia and cancer, expression in the cytoplasm predominates and may have anti- and pro-tumor action depending on the applied therapies associated with stimulation (Urban-Wojciuk et al., 2019). TLR3s, on the other hand, are more expressed in endosomes in antigen-presenting immune cells and epithelial cells as well as in multiple neoplasms, such as breast cancer, prostate cancer, and ovarian cancer, and may have anti- and pro-tumor action (Muresan et al., 2020). In both TLR2 and TLR3, the presence of interactions (hydrogen bonds, salt bonds and non-contact interactions) with the multi-epitope vaccine was identified.

Because of this, molecular dynamics were performed both in the case of the multi-epitope-TLR2 vaccine interaction, as well as in the multi-epitope-TLR3 vaccine to evaluate the behavior of the multi-epitope vaccine in a computer simulation with the equivalent biological conditions. The parameters used in both simulations revealed that the interactions of the multi-epitope vaccine with TLR3 are more stable than those presented with TLR2. Therefore, highlighting the multi-epitope vaccine's interaction with TLR3 as the most promising.

TLR3 stimulation can increase the production of type I IFNs, inhibit tumor cell proliferation, and stimulate immune cell antitumor phenotypes (Pahlavanneshan et al., 2021). Some clinical trials of TLR3 agonists are in phase II studies for colorectal cancer, melanoma, prostate cancer, breast cancer, head and neck squamous cell cancer, and non-Hodgkin's lymphoma, due to TLR3 ability to upregulate pathways that stimulate anti-tumor immune responses (Duan et al. 2022). These data reveal that TLR3 stimulation by agonists has antitumor potential and may be a therapeutic option in the treatment of cancer.

Finally, in silico cloning of the multi-epitope vaccine was possible by adapting the codons and inserting them into the linearized Allele TA vector. With this, it has been shown that the multi-epitope vaccine can be produced by cloning. Although there are advantages when working with linearized TA vectors for cloning, such as no need to prepare cohesive ends, the conformations (linear and circular) show different migration in the 1% agarose gel in different buffer solutions (Ishido; Ishikawa; Hirano, 2010). In this study, low ionic strength (SB) sodium boric acid buffer showed better resolution in 1% agarose gel migration.

On the other hand, the simulation of the immune response after the administration of the multi-epitope vaccine with one exposure and in three successive exposures at intervals of time, showed a favorable immune response. Being the vaccination schedule of more than one exposure to the multi-epitope vaccine showed a more lasting and effective response. The increase in NK cells and macrophages after each injection of the multi-epitope vaccine corroborates the fact that NK cells produce IFN-γ which activates macrophages and are co-stimulated to proliferate by IL-2 (Habanjar et al., 2023).

Regarding adaptive immunity, the Th1 lymphocyte subpopulation had a significant increase, which in turn induces cytotoxic T lymphocytes. In addition, there was a gradual increase in immunoglobulins, which indicates a humoral response of B lymphocytes. This humoral response is a result of the recognition of the multi-epitope vaccine by the immune system and may be a biomarker of the immune response in cancer patients (Astaneh; Dashti; Esfahani, 2019).

Thus, the multi-epitope vaccine showed wide population coverage, physicochemical properties that guarantee stability to the molecule and viability of application in biological systems, interactions between the Toll-like receptor, TLR-3, synthesis capacity by cloning and immune response.

In this study, the development of a multi-epitope vaccine with antitumor action based on the cancer-testis antigen TFDP3 was proposed. This multi-epitope vaccine showed wide population coverage, low homology with other proteins, relevant physicochemical parameters, ability to interact in biological systems, especially with TLR3, cloning capacity and stimulation of the immune response that revealed it as a potential candidate for immunotherapy of cancer types that express TFDP3. It should be noted that this is the first peptide-based vaccine designed against the cancer-testis antigen TFDP3 and that the evaluation was performed by screening vaccine epitopes using immunoinformatics tools. Therefore, future studies are needed to explore this multi-epitope vaccine in vitro and in vivo assays to corroborate the findings.

Acknowledgments

This research was supported by Federal University of Alagoas, Federal Institute of Alagoas, Oswaldo Cruz Foundation of Rondonia and Alagoas State Research Support Foundation.

Conflict of interest

The authors have no conflicts of interest to declare.

Author Contribution

G. C. O. N. and J. W. B. S. J. contributed to data analysis, G. C. O. N., J. W. B. S. J., C. S. M. and C. A. C. F. wrote the main manuscript text, G. C. O. N. and J. W. B. S. J. prepared tables and figures, G. C. O. N. and F. B. Z. molecular dynamics analyses, and G. C. O. N., R. M. L. R. , C. S. M., C. A. C. F., A. K. S. F. D. and E. S. B. G. reviewed the manuscript.

Abbott M, Ustoyev Y. Cancer and the Immune System: The History and Background of Immunotherapy. Seminars in Oncology Nursing. 2019 Oct 1;35(5):150923.
Aghajani J, Farnia P, Farnia P, Ghanavi J, Velayati AA. Molecular Dynamic Simulations and Molecular Docking as a Potential Way for Designed New Inhibitor Drug without Resistance. Tanaffos. 2022;21(1):1-14.
Astaneh M, Dashti S, Esfahani Z. Humoral immune responses against cancer-testis antigens in human malignancies. Human Antibodies. 2019 Nov 15;27(4):237–40.
Backert L, Kohlbacher O. Immunoinformatics and epitope prediction in the age of genomic medicine. Genome Medicine. 2015;7:119.
Baek M, DiMaio F, Anishchenko I, et al. Accurate prediction of protein structures and interactions using a three-track neural network. Science. 2021;373(6557):871-876.
Berendsen HJC, van der Spoel D, van Drunen R. GROMACS: A message-passing parallel molecular dynamics implementation. Comput Phys Commun. 1995;91(1-3):43-56.
Bahrami AA, Payandeh Z, Khalili S, Mahdavi M, Parvizi F, Zarei M, et al. Immunoinformatics: In Silico Approaches and Computational Design of a Multi-epitope, Immunogenic Protein. Int Rev Immunol. 2019;38(6):307-22.
Bhasin M, Raghava GPS. Analysis and prediction of affinity of TAP binding peptides using cascade SVM. Protein Sci. 2004;13(3):596-607.
Blom N, Gammeltoft S, Brunak S. Sequence and structure-based prediction of eukaryotic protein phosphorylation sites. J Mol Biol. 1999;294(5):1351-62.
Bui HH, Sidney J, Li W, Fusseder N, Sette A. Predicting population coverage of T-cell epitope-based diagnostics and vaccines. BMC Bioinformatics. 2006;7:153.
Casey SC, Amedei A, Aquilano K, Azmi AS, Benencia F, Bhutia SK, et al. Cancer prevention and therapy through the modulation of the tumor microenvironment. Semin Cancer Biol. 2015;35 Suppl:S199-S223.
Clark DP, Pazdernik NJ, McGehee MR. Cloning Genes for Synthetic Biology. In: Clark DP, Pazdernik NJ, McGehee MR, editors. Molecular Biology. 3rd ed. Academic Cell; 2019. p. 199-239.
Chu M, Zheng X, Zheng Y, Zhang H, Deng X, Zhang P, et al. TFDP3 confers chemoresistance in minimal residual disease within childhood T-cell acute lymphoblastic leukemia. Oncotarget. 2017;8(1):1405-15.
Colovos C, Yeates TO. Verification of protein structures: patterns of nonbonded atomic interactions. Protein Sci. 1993;2(9):1511-9.
Comeau SR, Gatchell DW, Vajda S, Camacho CJ. ClusPro: an automated docking and discrimination method for the prediction of protein complexes. Bioinformatics. 2004;20(1):45-50.
Desai R, Coxon AT, Dunn GP. Therapeutic applications of the cancer immunoediting hypothesis. Semin Cancer Biol. 2022;78:63-77.
Dhanda SK, Vir P, Raghava GP. Designing of interferon-gamma inducing MHC class-II binders. Biol Direct. 2013;8:30.
Diefenbach CS, Gnjatic S, Sabbatini P, Jungbluth AA, Ginsberg R, Williamson B, et al. Safety and immunogenicity study of NY-ESO-1b peptide and montanide ISA-51 vaccination of patients with epithelial ovarian cancer in high-risk first remission. Clin Cancer Res. 2008;14(9):2740-8.
Dimitrov I, Bangov I, Flower DR, Doytchinova I. AllerTOP v.2 - a server for in silico prediction of allergens. J Mol Model. 2014;20(6):2278.
Ding L, Wang L, Zhang Y, Zhao Q, Li X, Zhang W, et al. TFDP3 regulates the apoptosis and autophagy in breast cancer cell line MDA-MB-231. PLoS One. 2018;13(9).
Doytchinova IA, Flower DR. VaxiJen: a server for prediction of protective antigens, tumour antigens and subunit vaccines. BMC Bioinformatics. 2007;8:4.
Duan T, Duan S, Huang M, Liang L, Lu F, Li R. Toll-Like Receptor Signaling and Its Role in Cell-Mediated Immunity. Front Immunol. 2022;13:812774.
Eisenberg D, Lüthy R, Bowie JU. VERIFY3D: assessment of protein models with three-dimensional profiles. Methods Enzymol. 1997;277:396-404.
Escribese MM, Barber D. New insight into cancer immunotherapy. Allergol Immunopathol (Madr). 2017;45 Suppl 1:50-55.
Farran B, Nisbet LA, Carman BL, Shao Q, Kwok TT, Lopez DS, et al. Folate-targeted immunotherapies: Passive and active strategies for cancer. Cytokine Growth Factor Rev. 2019;45:45-52.
Garrido F, Aptsiauri N. Cancer immune escape: MHC expression in primary tumours versus metastases. Immunology. 2019;158(4):255-66.
Gasteiger E, Hoogland C, Gattiker A, Duvaud S, Wilkins MR, Appel RD, et al. Protein identification and analysis tools on the ExPASy server. In: Walker JM, editor. The Proteomics Protocols Handbook. Totowa, NJ: Humana Press; 2005. p. 571-607.
Goodswen SJ, Kennedy PJ, Ellis JT. A guide to current methodology and usage of reverse vaccinology towards in silico vaccine discovery. FEMS Microbiol Rev. 2023;47(2).
Grote A, Hiller K, Scheer M, Munch R, Nortemann B, Hempel DC, et al. JCat: a novel tool to adapt codon usage of a target gene to its potential expression host. Nucleic Acids Res. 2005;33.
Gupta S, Kapoor P, Chaudhary K, Gautam A, Kumar R, Open Source Drug Discovery Consortium, et al. In Silico Approach for Predicting Toxicity of Peptides and Proteins. PLoS One. 2013;8(9).
Haabeth OA, Lorvik KB, Fauskanger M, Hammarström C, Donaldson IM, Haraldsen G, et al. How do CD4+ T cells detect and eliminate tumor cells that either lack or express MHC class II molecules? Front Immunol. 2014;5:174.
Habanjar O, Diot A, Karaky R, Fakhoury I. Crosstalk of Inflammatory Cytokines within the Breast Tumor Microenvironment. Int J Mol Sci. 2023;24(8):4002.
Herrera LRM. In Silico Approach in Designing a Novel Multi-Epitope Vaccine Candidate against Non-Small Cell Lung Cancer with Overexpressed G Protein-Coupled Receptor 56. Asian Pac J Cancer Prev. 2020;21(8):2297-306.
Hess B, Bekker H, Berendsen HJC, Fraaije JGEM. LINCS: A linear constraint solver for molecular simulations. J Comput Chem. 1997;18(12):1463-72.
Huang J, Deng J, Wang X, Zhang H, Zhang P, Xu Z, et al. TFDP3 as E2F Unique Partner, Has Crucial Roles in Cancer Cells and Testis. Front Oncol. 2021;11:652759.
Hu Y, Lu L, Yu J, Zhang Y, Jin X, Wang Y, et al. Long-term outcomes of helper peptide vaccination for metastatic melanoma. Ann Surg. 2015;262(3):456-64.
Ishido T, Ishikawa M, Hirano K. Analysis of supercoiled DNA by agarose gel electrophoresis using low-conducting sodium threonine medium. Anal Biochem. 2010;400(1):148-50.
Jones DT. Protein secondary structure prediction based on position-specific scoring matrices. J Mol Biol. 1999;292(2):195-202.
Johnson M, Zaretskaya I, Raytselis Y, Merezhuk Y, McGinnis S, Madden TL. NCBI BLAST: a better web interface. Nucleic Acids Res. 2008;36.
Jorgensen WL, Chandrasekhar J, Madura JD, Impey RW, Klein ML. Comparison of simple potential functions for simulating liquid water. J Chem Phys. 1983;79(2):926-35.
Jorgovanovic D, Song M, Wang L, Zhang Y. Roles of IFN-γ in tumor progression and regression: a review. Biomark Res. 2020;8:49.
Keshavarz-Fathi M, Rezaei N. Cancer Immunology. Switzerland: Springer; 2018.
Kesmir C, Nussbaum AK, Schild H, Detours V, Brunak S. Prediction of proteasome cleavage motifs by neural networks. Protein Eng. 2002;15(4):287-96.
Kringelum JV, Lundegaard C, Lund O, Nielsen M. Reliable B cell epitope predictions: impacts of method development and improved benchmarking. PLoS Comput Biol. 2012;8(12).
Krishnamoorthy HR, Karuppasamy R. Design and In Silico Validation of a Novel MZF-1-Based Multi-Epitope Vaccine to Combat Metastatic Triple Negative Breast Cancer. Vaccines. 2023; 11(3):577.
Kumar S, et al. A candidate triple-negative breast cancer vaccine design by targeting clinically relevant cell surface markers: an integrated immuno and bio-informatics approach. 3 Biotech. 2022;12(3).
Lam RA, et al. Cancer-testis antigens in triple-negative breast cancer: Role and potential utility in clinical practice. Cancers (Basel). 2021;13(15):1-22.
Lasek W. Cancer immunoediting hypothesis: history, clinical implications and controversies. Cent Eur J Immunol. 2022;47(2):168-74.
Laskowski RA. PDBsum new things. Nucleic Acids Res. 2009;37.
Laskowski RA, MacArthur MW, Thornton JM. PROCHECK: validation of protein-structure coordinates. In: International Tables for Crystallography. 2012. p. 684-7.
Lee SJ, et al. A Potential Protein Adjuvant Derived from Mycobacterium tuberculosis Rv0652 Enhances Dendritic Cells-Based Tumor Immunotherapy. PLoS One. 2014;9(8).
Lee GR, Heo L, Seok C. Effective protein model structure refinement by loop modeling and overall relaxation. Proteins. 2015;84(S1):293-301.
Liu J. Cancer vaccines as promising immuno-therapeutics: platforms and current progress. J Hematol Oncol. 2022;15(1):1-26.
Livingston B, et al. A Rational Strategy to Design Multiepitope Immunogens Based on Multiple Th Lymphocyte Epitopes. J Immunol. 2002;168(11):5499-506.
Ma Y, et al. TFDP3 was expressed in coordination with E2F1 to inhibit E2F1-mediated apoptosis in prostate cancer. Gene. 2014;537(2):253-9.
Mahdevar E, Safavi A, Abiri A, Kefayat A, Hejazi SH, Miresmaeili SM, Iranpur Mobarakeh V. Exploring the cancer-testis antigen BORIS to design a novel multi-epitope vaccine against breast cancer based on immunoinformatics approaches. J Biomol Struct Dyn. 2022;40(14):6363-80.
Magnan CN, Randall A, Baldi P. SOLpro: Accurate sequence-based prediction of protein solubility. Bioinformatics. 2009;25(17):2200-7.
McCall KD, Muccioli M, Benencia F. Toll-Like Receptors Signaling in the Tumor Microenvironment. In: Tumor Microenvironment. Advances in Experimental Medicine and Biology. Cham: Springer; 2020. v. 1223.
McGuffin LJ, Bryson K, Jones DT. The PSIPRED protein structure prediction server. Bioinformatics. 2000;16(4):404-5.
Meng X, et al. A novel era of cancer/testis antigen in cancer immunotherapy. Int Immunopharmacol. 2021;98.
Muresan XM, et al. Toll-Like Receptor 3 in Solid Cancer and Therapy Resistance. Cancers (Basel). 2020;12(11).
Nezafat N, et al. A novel multi-epitope peptide vaccine against cancer: an in silico approach. J Theor Biol. 2014;349:121-34.
Nin DS, Deng LW. Biology of Cancer-Testis Antigens and Their Therapeutic Implications in Cancer. Cells. 2023;12(6).
Pahlavanneshan S, et al. Toll-Like Receptor-Based Strategies for Cancer Immunotherapy. J Immunol Res. 2021;2021:9912188.
Parvizpour S, et al. Epitope-based vaccine design: a comprehensive overview of bioinformatics approaches. Drug Discov Today. 2020;25(6):1034-42.
Parrinello M, Rahman A. Polymorphic transitions in single crystals: A new molecular dynamics method. J Appl Phys. 1981;52(12).
Patra P, et al. Identification and Design of a Next-Generation Multi Epitopes Bases Peptide Vaccine Candidate Against Prostate Cancer: An In Silico Approach. Cell Biochem Biophys. 2020;78:495-509.
Pearlman DA, et al. AMBER, a package of computer programs for applying molecular mechanics, normal mode analysis, molecular dynamics and free energy calculations to simulate the structural and energetic properties of molecules. Comput Phys Commun. 1995;91:1-41.
Pitt JM, et al. Targeting the tumor microenvironment: removing obstruction to anticancer immune responses and immunotherapy. Ann Oncol. 2016;27(8):1482-92.
Ponomarenko J, et al. ElliPro: a new structure-based tool for the prediction of antibody epitopes. BMC Bioinformatics. 2008;9:514.
Rahmani A, et al. Development of a conserved chimeric vaccine based on helper T-cell and CTL epitopes for induction of strong immune response against Schistosoma mansoni using immunoinformatics approaches. Int J Biol Macromol. 2019;141:125-36.
Rapin N, et al. Computational Immunology Meets Bioinformatics: The Use of Prediction Tools for Molecular Binding in the Simulation of the Immune System. PLoS One. 2010;5(4).
Reynisson B, et al. NetMHCpan-4.1 and NetMHCIIpan-4.0: improved predictions of MHC antigen presentation by concurrent motif deconvolution and integration of MS MHC eluted ligand data. Nucleic Acids Res. 2020;48(W1).
Rock KL, Reits E, Neefjes J. Present Yourself! By MHC Class I and MHC Class II Molecules. Trends Immunol. 2016;37(11):724-37.
Saha S, Raghava GP. Prediction of continuous B-cell epitopes in an antigen using recurrent neural network. Proteins. 2006;65(1):40-8.
Safavi A, et al. In silico analysis of transmembrane protein 31 (TMEM31) antigen to design novel multiepitope peptide and DNA cancer vaccines against melanoma. Mol Immunol. 2019;112:93-102.
Schrödinger L, Delano W. PyMOL. Available from: http://www.pymol.org/pymol. Accessed on: September 14, 2023.
SnapGene Software. Available from: https://www.snapgene.com/. Accessed on: November 27, 2023.
Sonpavde G, et al. HLA-restricted NY-ESO-1 peptide immunotherapy for metastatic castration resistant prostate cancer. Invest New Drugs. 2014;32(2):235-42.
Steentoft C, et al. Precision mapping of the human O‐GalNAc glycoproteome through SimpleCell technology. EMBO J. 2013;32(10).
Sufyan M, et al. Implementation of Vaccinomics and In-Silico Approaches to Construct Multimeric Based Vaccine Against Ovarian Cancer. Int J Pept Res Ther. 2021;27(4):2845-59.
Teng MWL, Kershaw MH, Smyth MJ. Cancer Immunoediting: From Surveillance to Escape. 2nd ed. [S.l.]: Elsevier; 2013.
Vieira IHP, et al. Dinâmica visual: uma aplicação WEB para simulação de dinâmica molecular usando GROMACS. BMC Bioinformatics. 2023;24:107.
Wang X, et al. Development and validation of epithelial mesenchymal transition-related prognostic model for hepatocellular carcinoma. Aging (Albany NY). 2021;13(10):13822-45.
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.
Wilson B, Geetha KM. Lipid nanoparticles in the development of mRNA vaccines for COVID-19. J Drug Deliv Sci Technol. 2022;74:103553.
Yin K, et al. TFDP3 Regulates Epithelial-Mesenchymal Transition in Breast Cancer. PLoS One. 2017;12(1).
Zarling AL, et al. Phosphorylated peptides are naturally processed and presented by major histocompatibility complex class I molecules in vivo. J Exp Med. 2000;192(12):1755-62.
Zhang L, et al. Peptide-based materials for cancer immunotherapy. Theranostics. 2019;9(25):7807-25.
Urban-Wojciuk Z, et al. The Role of TLRs in Anti-cancer Immunity and Tumor Rejection. Front Immunol. 2019;10:2388.
Vasan N, Baselga J, Hyman DM. A view on drug resistance in cancer. Nature. 2019;575:299-309.
Vazquez E, Corchero JL, Villaverde A. Post-production protein stability: trouble beyond the cell factory. Microb Cell Fact. 2011;10:60.
Wieczorek M, Abualrous ET, Sticht J, Álvaro-Benito M, Stolzenberg S, Noé F, Freund C. Major Histocompatibility Complex (MHC) Class I and MHC Class II Proteins: Conformational Plasticity in Antigen Presentation. Front Immunol. 2017;8:292.
Wang X, et al. Development and validation of epithelial mesenchymal transition-related prognostic model for hepatocellular carcinoma. Aging (Albany NY). 2021;13(10):13822-45.
Wieczorek M, Abualrous ET, Sticht J, Álvaro-Benito M, Stolzenberg S, Noé F, et al. Major Histocompatibility Complex (MHC) Class I and MHC Class II Proteins: Conformational Plasticity in Antigen Presentation. Frontiers in Immunology. 2017 Mar 17;8(1). ‌
World Health Organization (WHO). GLOBOCAN v. 2022. Cancer Incidence and Mortality Worldwide. Available from: https://gco.iarc.who.int/en. Accessed on: February 1, 2024.

No competing interests reported.

SupplementaryinformationTollLikereceptor3TLR3agonistsinamultipeptidevaccineforTFDP3expressingcancers.docx

Toll-Like receptor 3 (TLR3) agonists in a multi-peptide vaccine for TFDP3 expressing cancers

Status:

Version 1

Abstract

Figures

INTRODUCTION

MATERIALS AND METHODS

RESULTS

DISCUSSION

CONCLUSION

Declarations

Author Contribution

References

Additional Declarations

Supplementary Files

Status:

Version 1