Immunoinformatics Approach to Design a Chimeric CD70-Peptide Vaccine against Renal Cell Carcinoma

Renal cell carcinoma (RCC) accounts for the majority of cancer-related deaths worldwide. Overexpression of CD70 has been linked to advanced stages of RCC. Therefore, this study aims to develop a multiepitope vaccine targeting the overexpressed CD70 using immunoinformatics techniques. In this investigation, in silico multiepitope vaccines were constructed by linking specific CD70 protein epitopes for helper T lymphocytes and CD8+ T lymphocytes. To enhance immunogenicity, sequences of cell-penetrating peptide (CPP), penetratin (pAntp), along with the entire sequence of tumor necrosis factor-α (TNF-α), were attached to the N-terminal and C-terminal of the CD70 epitopes. Computational assessments were performed on these chimeric vaccines for antigenicity, allergenicity, peptide toxicity, population coverage, and physicochemical properties. Furthermore, refined 3D constructs were subjected to a range of analyses, encompassing structural B-cell epitope prediction and molecular docking. The chosen vaccine construct underwent diverse assessments such as molecular dynamics simulation, immune response simulation, and in silico cloning. All vaccines comprised antigenic, nontoxic, and nonallergenic epitopes, ensuring extensive global population coverage. The vaccine constructs demonstrated favorable physicochemical characteristics. The binding affinity of chimeric vaccines to the TNF receptor remained relatively stable, influenced by the alignment of vaccine components. Molecular docking and dynamics analyses predicted stable interactions between CD70-CPP-TNF and the TNF receptor, indicating potential efficacy. In silico codon optimization and cloning of the vaccine nucleic acid sequence were accomplished using the pET28a plasmid. Furthermore, this vaccine displayed the capacity to modulate humoral and cellular immune responses. Overall, the results suggest therapeutic potential for the chimeric CD70-CPP-TNF vaccine against RCC. However, validation through in vitro and in vivo experiments is necessary. This trial is registered with NCT04696731 and NCT04046445.


Introduction
Renal cell carcinoma (RCC), a type of kidney cancer, accounts for over 90% of kidney-associated malignancies [1].Managing RCC remains a significant challenge, particularly concerning its metastatic form [2]. The advancements in immunotherapy have led to the approval of drugs aimed at combating advanced RCC [1,[3][4][5][6].However, treatment outcomes exhibit considerable variability [4][5][6].Several factors contribute to this variability, including inappropriate adjuvants [3,4], tumor cell heterogeneity [5,6], antigen loss within tumors [7], diminished MHC expression [8], inadequate T cell infiltration in tumor tissues [9], and immune suppression via dysfunctional T cells [10].Notably, the clinical realm has witnessed a surge in antigenfocused T cell therapies for advanced RCC, leveraging immunotherapeutic techniques like peptide-based vaccines [11] to enhance antitumor effects while mitigating adverse impacts.Designing a vaccine for RCC presents multifaceted challenges rooted in the complex nature of this malignancy [12].Tumorspecific antigens and immune evasion strategies by cancer cells hinder effective immune response elicitation.Additionally, the tumor microenvironment's immunosuppressive nature often thwarts vaccine-induced immune reactions [12].Moreover, the variability in tumor mutational landscape and the interplay of host factors further complicate efforts to devise a universally effective vaccine.These challenges necessitate innovative strategies to surmount the hurdles in creating a potent and broadly applicable vaccine against RCC.Overlapping or multiepitope peptide sequences recognized by both CD8 + and CD4 + T cells may help to overcome challenges like tumor heterogeneity, tumor antigen downregulation, and varied HLA haplotypes [4].
Several clinical investigations have aimed at distinct tumorassociated antigens in individuals with RCC.While these studies demonstrated certain favorable advancements, their outcomes revealed only partial responses when compared to the existing treatments [3][4][5][6][7][8][9][10][13][14][15].Lately, there has been a growing focus on directing efforts toward the CD70 molecule in RCC.CD70 serves as a tumor-associated antigen, exhibiting high expression across various malignancies [16][17][18], such as RCC [19].CD70, a type II transmembrane protein belonging to the TNF family, is typically expressed on antigenpresenting cells.However, in T and B cells, this molecule is inducible and becomes expressed upon activation [20].The interaction between CD70 and its corresponding receptor CD27 plays a crucial role in modulating the immune response, regulating activation, and governing proliferation after antigen stimulation [21].Studies have suggested that the presence of CD70 in RCC tumors correlates with the reduced survival rates [20].This observation underscores the rationale behind choosing CD70 as a suitable target for RCC immunotherapy.A current clinical trial (NCT04696731) is assessing the safety and effectiveness of CAR-T cell therapy directed against CD70 in RCC [20,22,23].
Research into peptide-based cancer vaccines has been thorough, yet their immunogenicity remains moderate.Efforts are underway to explore various strategies aimed at enhancing the efficacy of these vaccines [24].One way to tackle the inadequacy in the immune response is by combining an adjuvant with the vaccine antigen [25,26].Cytokines have emerged as promising adjuvants due to their safe biomaterial nature, derived from humans, and their ability to modulate the immune system [27,28].Tumor necrosis factor-α (TNF-α) is a critical proinflammatory cytokine primarily produced by macrophages and T lymphocytes, playing a pivotal role in both the innate and adaptive immune responses [29].TNF-α cytokine plays a significant role in multiple stages of the immune process, including the activation of the innate immune system, maturation and recruitment of dendritic cells (DC), facilitation of T cell activation, and clearance of pathogens [28].Many studies across various model systems have highlighted the adjuvant properties of TNF-α in combating viral infections [30][31][32].One explored method to enhance the efficacy of peptide vaccines involves conjugating antigens with cell-penetrating peptides (CPPs).CPPs are linear peptides possessing cationic and amphiphilic traits.Studies suggest that antigen-CPP conjugates have a heightened capacity to stimulate the immune system compared to the unbound peptides [24].CPPs aid in facilitating the uptake of the vaccine into the lymphatic system and extending its life span in vivo [24].Studies conducted in vitro have demonstrated the ability of CPPs to promote the uptake of antigen peptides by DCs, elevate vaccine immunogenicity in animal models like non-human primates, and improve the efficacy of cancer vaccines in mice, as supported by multiple research endeavors [33][34][35][36][37][38][39][40][41].Current clinical trials (clinicaltrials.gov:NCT04046445) are actively evaluating the efficacy of antigen-CPP vaccines.In vitro studies have highlighted that coupling with the potent cell-penetrating peptide penetratin (pAntp) enhances antigen uptake and presentation by DCs [42,43].Based on these rationales, this study aims to develop a peptide vaccine incorporating CD70, TNF-α and CPP penetratin.

Prediction of Immunogenic Epitopes of CD8
+ and CD4 + T Cells.The web server NetMHCpan-4.1 (https://services.healthtech.dtu.dk/service.php?NetMHCpan-4.1)was configured to predict 9 mer through 14 mer epitopes for HLA class I alleles that were expressed by CD70.The default threshold parameters of 0.5 and 2 were applied to categorize high and low affinity for MHCI binders.The NetMHCpan-4.1 predicts epitopes for 12 HLA class I supertypes, wherein each supertype denotes a group of closely associated HLA alleles with shared affinities for specific peptides due to similar structural traits in the HLA peptide-binding grooves [26].The web version of NetMHCIIpan-4.0 (https://services.healthtech.dtu.dk/service.php?NetMHCIIpan-4.0)was utilized to predict MHCII binders.By default, epitopes with a length of 15 mers were chosen, and thresholds of 1 and 5 were applied to distinguish high and low affinities, respectively.
2.5.Population Coverage Analysis.Ensuring widespread population coverage is another crucial element in designing vaccines [46].Variations in HLA types occur among global populations, influencing the antigens presented to HLA molecules.To evaluate the vaccine's coverage across diverse populations worldwide, the IEDB resource's population coverage analysis module (http://tools.iedb.org/population/),with default parameters was utilized.This analysis encompassed MHC-I and -II coverage across different regions, countries, and ethnic groups.
2.6.Peptide-Protein Flexible Docking.Assessing the potential formation of MHC-peptide complexes was done using the GalaxyPepDock server (http://galaxy.seoklab.org/cgi-bin/submit. cgi?type=PEPDOCK).Specifically, the binding interactions of each selected CD70 epitope with various MHC alleles were examined.The PDB IDs corresponding to HLA alleles, sourced from the RCSB database (https://www.rcsb.org),were listed in Tables 1 and 2.
2.7.Multiepitope Peptide Constructs.We employed topscoring CTL and Th epitopes to create the final multiepitope vaccine designs.These epitopes were connected via the AAY proteolytic linker sequence to construct chimeric forms.Six peptide constructs were developed utilizing SnappGene ® 3.2.1,encompassing T cell epitopes sourced from CD70, linked to both the N-terminal and C-terminal regions of the CPP epitope, along with the complete sequence of TNF-α, as an adjuvant.Additionally, a construct containing the entire sequence of TNF-α was established as a reference for comparative analysis with the other constructs.2.11.Vaccine-Receptor Docking and Interaction.Molecular docking involves the interaction of a ligand with its receptor, resulting in the creation of a stable complex [50].Additionally, it can predict the binding strength between two molecules using specific scoring functions.For the docking simulations, we selected the TNF receptor and paired it with the optimized vaccine model.The 3D structure of the TNF receptor was created, refined, and validated using the servers outlined in Section 2.9.The ClusPro (https://cluspro.bu.edu/) server was utilized to evaluate the binding affinity between the multiepitope vaccines and the TNF receptor through molecular docking [51].To scrutinize the interactions within the docked complex, DIMPLOT within LigPlot + v2.2 was utilized to analyze the interplay among residues of Chain-A (TNF receptor) and Chain-B (vaccine construct) [52].The input was a PDB file depicting the vaccine-receptor complex.The output provided an in-depth representation of the intermolecular interactions, such as hydrogen bonds.Dotted lines in green and red visually depicted hydrogen bonds and salt bridges, respectively.
2.12.Normal Mode Analysis.To assess the macromolecular mobility and stability of the selected vaccine construct bound to the TNF receptor, we utilized the iMODS server (http:// imods.chaconlab.org)[53].We chose this particular server due to its superior efficiency and speed compared to other MD simulation methods available [54].The iMODS server creates protein models that connect atoms using harmonic springs and employs normal mode analysis (NMA) to study the collective movement of proteins [55].

Molecular Dynamics (MD) Simulation of Vaccine
Constructs-Receptor Complexes.GROMACS 2020 and CHARMM36m force field conducted molecular dynamics (MD) simulations that model extensive biomolecular systems across time using efficient parallel algorithms [57].
The GROMACS software facilitated a 100 ns simulation of the docked complex involving the vaccine construct and TNF receptor.The GROMACS solvate tool was utilized to envelop the protein surface with a TIP3P water layer, spanning 10 Å. Sodium and chloride ions were introduced to neutralize the surface charges of the structure after the water layer addition.The system's atom count was verified, confirming a total of 407,960 atoms.To prevent undesired interactions between water and the complex, the system's energy was minimized using the steepest descent algorithm.Subsequently, the system underwent gradual heating from 0 to 300 K under constant volume for 200 ps, followed by equilibration at a consistent pressure.Throughout the 100-ns period, parameters including root-mean-square deviation (RMSD), root-mean-square fluctuation (RMSF), hydrogen bond count (H-bonds), radii of gyration (Rg), solvent accessible surface area (SASA), and ligand-receptor interaction energy were continuously computed.
2.14.Codon Optimization and In Silico Cloning.We employed a codon optimization strategy to enhance the expression of the recombinant protein.Given the degeneracy in the genetic code, where most amino acids can be encoded by multiple codons, optimizing codon usage becomes crucial.The Java Codon Adaptation Tool (JCat) server (http://www.prodoric.de/JCat) was utilized, specifically in the Escherichia coli (strain K12) codon system, to determine the codon adaptation index (CAI) values and GC contents, crucial indicators for assessing protein expression levels [58].An optimal CAI value is 1.0, with scores above 0.8 considered favorable, while maintaining a GC content within the 30%-70% range.Translation and transcriptional efficiencies are adversely affected beyond these limits [59].Subsequently, to confirm vaccine expression, the optimized sequence of the final vaccine construct, inclusive of restriction sites, was inserted into the E. coli plasmid vector pET−28 a (+) using SnapGene software (https://www.snapgene.com/free-trial/).

Immune Simulations.
The C-ImmSim online server (https://kraken.iac.rm.cnr.it/C-IMMSIM/) was utilized to anticipate the immune response triggered by the vaccine.
Employing the Celada-Seiden model, C-ImmSim portrays the immune system's dual-arm profile following exposure to the vaccine construct [60].Simulations involved administering the vaccine without LPS and with three injections spaced at intervals of 1, 84, and 168.All other parameters were maintained at their default settings.

Results
3.1.The Sequences of the CD70, TNF-α Proteins, and CPP.The human CD70 and TNF-α reference sequences were sourced from NCBI and processed in FASTA format for subsequent analyses.Additionally, the CPP peptide sequences were obtained from a previous study for further use [24].

CTL Epitope Prediction.
The CD70 protein sequence underwent analysis using the NetMHCpan 4.1 server to identify its most immunodominant regions.This involved assessing these regions against high-frequency HLA-I alleles.
Notably, an inverse relationship between average rank scores and the binding affinity of epitopes to MHC-I alleles was observed.Higher binding affinity scores across multiple MHC-I alleles indicated superior CTL-stimulating epitopes.
From the predicted epitopes, those displaying better binding affinity were chosen, outlined in Table 3.It is worth noting that all the selected epitopes were confirmed to be nonallergenic, nontoxic, and exhibited antigenicity (Table 3).

T Cell CD4
+ Epitope Prediction.In the beginning, the exploration for potential CD70 epitopes in NetMHCIIpan-4.0 resulted in several sequences identified in humans.Following specific selection criteria, we refined the collection to two candidate CD70 epitopes for MHCII (Table 4).Considering their antigenic, allergenic, and toxic properties, the predicted CD70 epitopes were confirmed as nonallergenic, nontoxic, and exhibited antigenicity (Table 4).

MHC-I Processing and Immunogenicity of CTL Epitopes.
Table 5 displays the proteasome score, TAP score, epitope identification score, and immunogenicity score for each chosen MHCI epitope.Notably, all epitopes listed in Table 5 exhibited high-quality proteasomal cleavage and efficient TAP transport.

Population Coverage.
The top scoring epitopes and their respective population coverage percentage for human were indicated in Figure 1.The CD8 + and CD4 + T-cell epitopes showed a worldwide cumulative population coverage of 92%.Journal of Immunology Research [61].When triggered, T cells and NK (natural killer) cells produce IFN-γ, enhancing macrophage activation and antiviral mechanisms, while also augmenting antibody production [61].Table 4 displays the cytokine secretion potential of the screened HTL epitopes in humans to induce IFN-γ.Positive SVM scores in IFN-γ induction were observed for all selected CD70 epitopes.
3.7.Peptide-Protein Flexible Docking.Upon assessing the interaction of MHC-peptide pairs, the top ten complexes were examined, and among them, model 0 displayed the most favorable lowest energy.Consequently, due to its superior interaction similarity scores between the CD70 epitopes and MHC-I and II molecules, model 0 was chosen and detailed in Tables 1 and 2.
3.8.Design of Multiepitope Constructs.In this investigation, we employed the SnapGene ® 3.2.1 tool to design the final constructs, integrating the screened CTL and HTL epitopes sourced from the CD70 protein.Through this tool, we devised six structures encompassing the entire TNF protein alongside the CD70 and CPP epitopes.These constructs were arranged sequentially, connected by AAY linkers.Furthermore, the components of each vaccine construct were positioned adjacently, either at the N-terminal or C-terminal regions (CD70-CPP-TNF, CD70-TNF-CPP, CPP-CD70-TNF, CPP-TNF-CD70, TNF-CD70-CPP, and TNF-CPP-CD70).For visual reference, Figure S1 contains 2D representations of each construct.
3.9.The Physicochemical Parameters.The ProtParam server results were illustrated in Table 6.According to the data, the ideal molecular weight of the vaccine constructs falls within the range of 40-70 kDa [62].Surprisingly, our developed vaccine constructs exhibit a fitting molecular weight of 45,530.11Da for a vaccine construct, alongside a theoretical PI of 9.5, indicating the protein's basic characteristics.The protein instability index for the human constructs was calculated at 43.94.Moreover, the aliphatic index derived from the vaccine construction measured 91.22, while the GRAVY value stood at −0.166, suggesting a stable protein.The negative GRAVY value underscores the protein's hydrophilic nature [63].Based on the solubility value of soluble E. coli protein, our vaccine constructs were anticipated to have low solubility [61].Additionally, the constructs demonstrated antigenicity and were nonallergenic.
3.10.Tertiary Structure Prediction, Refinement, and Validation of 3D Structures.The 3D models of the chimeric vaccine constructs generated by the Robetta server displayed favorable Ramachandran plot statistics.Across all constructs, over 82% of the residues was located within the favored and additionally allowed regions on the Ramachandran plot.The top 3D models were further enhanced using the GalaxyRefine 2 server, resulting in improved tertiary structure quality.Tools like ERRAT2, Verify3D, and PROCHECK hold high regard for evaluating the quality of protein structures [64][65][66].Therefore, we performed an external quality assessment on the representative models using these tools.ERRAT2 calculates an overall quality factor representing the percentage of the protein structure with an error value below 95.This metric depends on the statistical distribution of pairwise atomic interactions, where a higher distribution of nonbonded atoms implies potential structural inaccuracies.The ERRAT score for the CD70-CPP-TNF vaccine stood at 86.41%, denoting good quality and validity (Figure 2).A score above 50 suggests a well-structured model [67].Verify3D examines the compatibility between 3D models and their amino acid sequences (1D).Scores below 80% indicate that less than 80% of the amino acids in the structures achieved a score ≥0.1 in the 3D/1D profile.Specifically, the CD70-CPP-TNF model displayed   Journal of Immunology Research 70.42% of residues with a 3D−1D score ≥0.1 (Figure 2).The vaccine constructs underwent validation via ProSA-web and Ramachandran plots postrefinement.The ProSA-web server yielded a Z-score of − 7.62 for the CD70-CPP-TNF chimeric vaccine (Figure 2), well within the acceptable range.In the Ramachandran plot, 89.5% of the residues were located in the most favored regions for this construct (Figure 2).These assessments collectively confirm the high quality of the final vaccines' 3D structures.Figures S2-S5 include the tertiary structures, ProSA-web z-score plots, Ramachandran plots, ERRAT, and verify 3D plots for the other chimeric constructs and TNF receptor.

Prediction of Structure-Based Epitopes on Vaccine
Constructs.The epitope sequence should allow sufficient exposure and projection for B-cell receptors (BCRs) to bind effectively.Ellipro (http://tools.iedb.org/ellipro/)predicts both linear and discontinuous B-cell epitopes by utilizing the PI of a residue.This server was utilized to identify structural epitopes in the refined and validated 3D constructs.Tables 7 and 8 display the properties of the predicted linear and discontinuous B-cell epitopes for the CD70-CPP-TNF vaccine construct, respectively.Tables S1 and S2 present the predicted linear and conformational B-cell epitopes for other constructs.The scores and the number of epitopes obtained from the Ellipro server suggest that the predicted linear and discontinuous epitopes could potentially trigger B cell immune responses.
3.12.Protein-Protein Docking between TNF Receptor and Multiepitope Peptide Constructs.The ClusPro 2.0 software facilitated the simulation and analysis of potential interactions between the ligands of each vaccine construct and the TNF receptor.In total, seven models (CD70-CPP-TNF, CD70-TNF-CPP, CPP-CD70-TNF, CPP-TNF-CD70, TNF-CD70-CPP, TNF-CPP-CD70, and TNF-α whole sequence) were generated for each docking process.Models with the lowest energy scores were chosen.Notably, there exists an inverse relationship between energy levels and binding affinity; in simpler terms, lower energy levels indicate higher binding affinity between constructs and receptor in docked complexes.Intriguingly, when comparing the binding score of the TNF-α whole sequence with the TNF receptor, the vaccine constructs displayed higher docking scores with this receptor (Table 9).This suggests that the alignment and composition of the chimeric vaccines might enhance their interaction with the TNF receptor.To be more specific, the energy levels of docking between vaccine constructs and the TNF receptor could be overshadowed by the alignment of vaccine components.Interestingly, the CD70-CPP-TNF vaccine construct showed the minimum energy level of docking.The interacting amino acids between the receptor and ligand in the docked complexes were visualized using the DIMPLOT module in LigPlot+ as depicted in Figure S6.

Stability Prediction of the Vaccine-Receptor Complexes Molecular Dynamics (MD).
The iMOD server was employed to forecast the structural stability of the CD70-CPP-TNF-TNF-receptor complex, which displayed the highest docking scores.Illustrated in Figure 3(a) is the docked complex of CD70-CPP-TNF-TNF-receptor. In Figure 3   complexes, revealed that the CD70-CPP-TNF-TNF receptor (4.5 × 10 −7 ) demanded considerable energy for complex deformation (Figure 3(d)).Generally, deformation becomes more difficult when the eigenvalue is high [55].Furthermore, residue-residue contacts in the covariance matrix were visualized in red, while uncorrelated and anticorrelated motions were indicated in blue and white, respectively (Figure 3(e)).
The elastic network model delineated atoms connected by springs, with stronger connections depicted by darker shades of gray (Figure 3(f )). .The results suggest a progressive increase in the number of hydrogen bonds formed between the chains, affirming the stability and strong connection within the complex.Assessing the complex stability further, the Rg values were tracked across the MD trajectories.Rg measures the elastic stability of a cross-section.The average Rg value for the vaccine during the simulation was determined to be 5.26.Initially, the vaccine displayed a Rg of approximately 6.5 nm, which gradually decreased to 5 nm over the course of the simulation (Figure 4(e)).Conversely, the average Rg value for the TNF receptor during the simulation was found to be 2.41 nm.The TNF receptor consistently maintained low Rg values, remaining stable throughout the entire simulation period (Figure 4(f)).Figure 4(g) depicted a notable rise in the TNF receptor's SASA over the simulation duration, while the SASA values of the vaccine remained constant, indicating the stability and compactness of its structure until the end of the simulation.The substantial SASA of the TNF receptor-vaccine complex indicates its stability within the solvent environment.Subsequently, we evaluated the nonbonded interaction energies-Lennard-Jones short-range (LJ-SR) and Coulombic short-range (Coul-SR) potential-between the TNF receptor and the vaccine (Figure 4(h)).Coul-SR accounts for the electrostatic interactions between the TNF receptor and the vaccine construct, while LJ-SR represents the van der Waals interactions between the two.Notably, the dominant interaction contributing to the stabilization of the TNF receptor and vaccine complex primarily arises from the electrostatic component, suggesting its critical role in binding.

Codon Optimization and
In Silico Cloning.We evaluated the cloning and expression efficiency of the vaccine.Additionally, we optimized the codon usage of the vaccine construct for E. coli strain K12.The JCat was employed for this purpose.The CAI value for the CD70-CPP-TNF construct was 0.960, with a mean GC proportion of 55.90%.These results suggest a high probability of successful expression of the vaccine in E. coli.After optimization, the codon sequences were incorporated into the pET28a (+) vector for cloning using SnapGene software.In Figure 5, the codon sequence of the final CD70-CPP-TNF vaccine is highlighted in red, representing the 1233 bp gene sequence generated by the JCat server.The pET28a (+) expression vector is illustrated in black.The codon sequence is inserted into the region between EcoRI and BamHI, resulting in a clone with a total length of 6596 bp.
3.16.Immune Simulation.The immunogenic profile depicted in Figure 6 shows the response to the CD70-CPP-TNF multipeptide vaccine.Notably, three immunizations at 4-week intervals elicited a rapid and robust immune reaction.The initial CD70-CPP-TNF vaccine response exhibited heightened levels of IgM, followed by more robust secondary and tertiary responses characterized by elevated levels of IgM + IgG, IgG1 + IgG2, IgG1 antibodies, and a swift decline in antigen concentration (Figure 6

Discussion
The rise of RCC poses a significant global public health threat, leading to heightened rates of mortality and morbidity.
Researchers have been motivated to develop an efficient treatment in response to this situation.Vaccination stands as a widely utilized method to bolster the host's immune response during disease scenarios [68].Immunoinformatic offers advantages in vaccine design by allowing rapid screening of potential epitopes, predicting immunogenicity, and aiding in the selection of the most effective candidates for vaccine development [69,70].Chimeric or multivalent antigens, which link potentially immunogenic molecules into a unified structure, present an appealing approach [71].For decades, researchers have dedicated their efforts to discovering a potent antigen and formulation for an RCC vaccine, resulting in the identification of promising antigens and a shared perspective on the significance of vaccination in tackling the disease [72].A recent study by Panowski et al. [73], demonstrated the development of CAR-T cells directed at CD70 single-chain antibodies.These anti-CD70 CAR-T cells demonstrated robust antitumor effects against RCC.This discovery underscores the potential of CD70 CAR-T cells in RCC treatment, leading to an ongoing Phase I clinical trial assessing the efficacy of CD70-CAR-T cells in metastatic RCC.Consequently, our main objective in this current study was to design a multiepitope vaccine encompassing T and B-cell epitopes derived from CD70 using immunoinformatics analysis.Extensive evidence supports the ability of antigenic epitopes to effectively trigger specific immune responses targeting the complete antigen [74].We employed NetMHCpan 4.0 and NetMHCIIpan 4.1 servers to uncover numerous potential epitopes within the CD70 protein.Epitopes were filtered based on their antigenic characteristics.Notably, none of the selected epitopes have been found in any studies.The chosen epitopes showed superior scores in MHC binding and extensive coverage across various MHCI and MHCII  subtypes.Efficient antigen processing and optimal presentation to the immune system are vital for a strong CD8 + T cell response.Hence, we assessed T cell antigen processing using the NETCTLpan1.1 server in this study.Each MHCI epitope exhibits a high-quality proteasomal cleavage and efficient TAP transport.Higher processing scores denote enhanced antigen processing outcomes [75,76].Given IFN-γ's pivotal role in acquired and innate immunity against tumors [77], we proceeded to assess the IFN-γ inducing HTL epitopes.All chosen epitopes were proficient in inducing IFN-γ.Due to the importance of HLA specificity in T-cell epitope selection [77], we employed molecular docking simulations to establish the connections between these epitopes and their respective HLA alleles.The analyses revealed robust binding in the epitope-MHC complexes, evident from their higher global energy scores.
Combining individual peptides into chimeric formations diminishes the risk of degradation by endo-or exopeptidases and augments their antigenicity, considering that individual peptides tend to have low immune-stimulating properties [78].In the pursuit of heightened antigenicity, the highest ranked epitopes were united using a suitable linker, culminating in the structure of the chimeric vaccine.Additionally, the final vaccine structure incorporated TNF-α adjuvant and CPP peptide to bolster the elicited immune response [79].In a multiepitope vaccination strategy, each epitope functions as an individual immunological unit.Consequently, careful scrutiny of their arrangement and placement within the designed constructs is imperative [80].Six multiepitope vaccines-CD70-CPP-TNF, CD70-TNF-CPP, CPP-CD70-TNF, CPP-TNF-CD70, TNF-CD70-CPP, and TNF-CPP-CD70, were developed.These constructs incorporated seven MHC class-I and two MHC class-II epitopes, TNF-α as an adjuvant, and the CPP peptide.The TNF-α whole sequence and CPP were strategically positioned at the C-terminal and N-terminal regions of the immunogenic CD70 epitopes, connected by the AAY linker.
Given the potential variance in the immunogenicity of individual epitopes when combined, it was crucial to assess the antigenicity of the designed constructs.Results from Vax-iJen v2 affirmed the antigenicity of all constructs.A proficient vaccine should not only evoke strong immune responses but also maintain favorable physicochemical and structural properties during production.Evaluation of the construct's physicochemical characteristics indicated their compliance with acceptable standards.Additionally, the assessment affirmed that the constructs posed no allergenicity concerns, ensuring the vaccine's safety profile.
Ensuring the quality of the tertiary structure in the vaccine construct is crucial as it significantly influences the presentation of peptides that trigger immune responses [81].The Robetta server was used for initial 3D modeling of the constructs, followed by refinement through Galaxy Refine, resulting in validated output models.The refinement notably elevated the tertiary structure of the vaccine to the desired standard.The ElliPro server predicted several linear and conformational B cell epitopes, indicating the constructs' potential to trigger immune responses encompassing both cellular and humoral aspects.
It is crucial to highlight that for the vaccine to exert its adjuvant effect, it must possess the capability to bind to the TNF receptor.The activation of innate immune cells through the TNF receptor is integral to the stimulation of cells in the adaptive immune system.In this context, the docking analysis  Journal of Immunology Research of the TNF receptor was examined.Our findings reveal novel peptide vaccines that interact effectively with TNF receptors, signifying satisfactory vaccine uptake by antigen-presenting cells (APCs).It is worth noting that the docking energy between the vaccine constructs and receptors might be impacted by how the vaccine components align.Specifically, TNF-α linked to CPP or CD70 epitopes demonstrated stronger docking with the TNF receptor compared to TNF-α alone.In general, CD70-CPP-TNF exhibited encouraging outcomes regarding its docking with the TNF receptor.Fascinatingly, studies have demonstrated that CPPs effectively transport proteins and peptides across various cell types [82].Therefore, incorporating CPPs into our vaccine design significantly enhanced the uptake of the vaccines by APCs.
The NMA data indicated that the docked proteins exhibit minimal deformability, suggesting a robust and stable binding capability of the vaccine-immune cell receptor complex.The anticipated eigenvalue of the vaccine-TNF receptor complex signifies its stability.Both the covariance and elastic network diagrams portray regions with correlated motions among amino acids and inflexible segments (signified by stiffer springs), respectively.The existence of these elements confirms the overall stability of the vaccine-receptor complex [83].
To assess the stability of the CD70-CPP-TNF-TNF receptor complex, molecular dynamics simulations were conducted.The RMSD results indicate the overall stability of the complex.Peaks in the graph represent highly flexible regions, while moderate fluctuations indicate relatively rigid areas.Based on Figure 4(c), there is evident fluctuation noticed within residues 1-170, which corresponds to the linker regions in the multiepitope vaccine structure.Following that, our focus shifted to examining the quantity of hydrogen bonds within the complex of the TNF receptor and the vaccine construct, which exhibited a gradual increase over the course of the simulations.This analysis further suggests the establishment

20
Journal of Immunology Research of a stable interaction between the TNF receptor and the vaccine structure.Moreover, the Rg graphs for both the vaccine and receptor proteins demonstrate that the TNF receptor displays a lower average Rg fluctuation value compared to the vaccine.The higher SASA value observed for the vaccinereceptor complex indicates its stability within the solvent.Additionally, the energy assessment revealed that the electrostatic interaction between the TNF receptor and vaccine contributes significantly to stabilizing the complex.In the process of developing a multiepitope vaccine, a crucial stage involves effective cloning and expression within a compatible vector [84].Given the redundancy in the genetic code, necessitating multiple codons to encode one amino acid [59], we carried out codon optimization and simulated cloning.This yielded enhanced expression levels and efficient translation of the vaccine when using pET-28a (+).Since, the vaccine comprises CTL and HTL epitopes, it has the potential to activate corresponding immune cells within the host.This activation could subsequently initiate signaling pathways that stimulate other immune cells [79].The counts of helper and cytotoxic T cells notably increased after the initial dose and continued to rise following subsequent doses.B cells also mirrored this trend: the primary response showed increased IgM levels, while the secondary and tertiary responses surpassed the primary one, displaying higher levels of IgM + IgG, IgG1 + IgG2, IgG1 antibodies, and a swift decrease in antigen concentration.The simulation using the construct prompted the production of IFN-γ and IL-2, supported by a suitable Simpson index (D) reflecting diversity.Moreover, the CD70-CPP-TNF construct triggered a minimal amount of IL-10, suggesting its potential to stimulate a TH1-mediated immune response, crucial for immunity against RCC.In sum, the designed vaccine holds significant promise from an immunological standpoint.Its immunogenicity stems from the selection of epitopes with high binding affinities to MHC class I and II alleles, crucial for robust T cell responses.The inclusion of CPPs and TNF-α sequences in the vaccine enhances antigenicity and aids in immune cell uptake, triggering a cascade of immune reactions.Moreover, the predicted B-cell epitopes suggest elicitation of a potent humoral response, crucial for longterm immunity.

Conclusion
In this research, we utilized computer-based techniques to create a multiepitope vaccination for CD70 as prophylaxis to human RCC.To the best of our knowledge, this study is the first to design potential vaccines against CD70-overexpressing RCC.Our vaccine design displayed considerable immune response.Nevertheless, additional examinations both in vitro and in vivo are necessary to confirm its actual capacity to trigger an effective immune response concerning RCC.However, translating in silico findings into experimental contexts comes with limitations and potential challenges.The reliance on algorithms for predictive accuracy and immunogenicity might not fully encompass the intricacies of the immune system.Also, biological variability, encompassing genetic, environmental, and health-related factors, could impact vaccine efficacy across different individuals.In addition, determining the long-term effectiveness and durability of the immune response induced by the designed vaccine poses challenges.In silico models may not precisely predict the longevity and robustness of the immune memory.Further, the practical manufacturing and scalability of a vaccine based on in silico results might face unexpected hurdles like production scalability, stability, and cost-effectiveness.Overcoming these challenges demands a multidisciplinary approach involving computational biologists, immunologists, clinicians, regulatory bodies, and industry partners [85].

2. 3 .
MHC-I Processing Prediction.The NetCTLpan 1.1 server (https://services.healthtech.dtu.dk/services/NetCTLpan-1.1/) was employed to assess the essential stages in MHC class I processing.NetCTLpan 1.1 processes protein sequences as input in 8-11 mer peptides, analyzing their binding with MHC class I, assessing proteasomal C-terminal cleavage, and considering TAP (transporter associated with antigen processing) transport.The threshold default values for the parameters were applied in the current study.2.4.Selection of Cytokine Inducing Epitopes.IFN-γ holds a critical role in initiating cellular immunity and subsequently 2 Journal of Immunology Research

1 FIGURE 1 :
FIGURE 1: Population coverage analysis.Prediction of the population coverage for six potential human vaccine candidates with MHC Class I and II alleles around the world.

FIGURE 3 :
FIGURE 3: The iMOD server's normal mode analysis of the human CD70-CPP-TNF-TNF receptor complex with normal mode analysis (NMA) outputted the following plots.(a) The docked complex of the CD70-CPP-TNF-TNF receptor complex.Stability of the protein-protein complex was investigated through.(b) B-factor values, (c) deformability, (d) eigenvalue, (e) covariance map (red, white, and blue colors correspond to correlated, uncorrelated and anticorrelated motions), and (f ) elastic network.The darker gray colors correspond to stiffer spring.

FIGURE 4 :
FIGURE 4: Based on the 100 ns MD simulations of the CD70-CPP-TNF-TNF receptor complex, the figure shows the: (a) root-mean-square deviation (RMSD) plot of vaccine.(b) RMSD plot of receptor.(c) Root-mean-square fluctuation (RMSF) plot of the ligand (blue line) and receptor (pink line).(d) Hydrogen bonds.(e) The radius of Gyration (Rg) plot of ligand.(f ) The Rg plot of receptor.(g) Solvent accessible surface area (SASA) of the vaccine construct (red line), TNF receptor (blue line), and vaccine-TNF receptor complex (black line).(h) Interaction energy plot over the simulation timescale.

FIGURE 5 :
FIGURE 5: In silico cloning of human CD70-CPP-TNF vaccine construct into pET28A (+) vector.The red region denotes the 1233 bp cloned vaccine nucleic acid sequence.

FIGURE 6 :
FIGURE 6: In silico simulation of immune response using CD70-CPP-TNF vaccine as an antigen after subsequent three injections.(a) Immunoglobulins and antigen levels, (b) B-cell population per state, (c) B-cell population, (d) T helper cell population per state, (e) T helper cell population, (f ) T cytotoxic cell population per state, and (g) production of cytokines.

TABLE 1 :
Interaction similarity scores of the selected CTL epitopes in CD70 protein.
4Higher rate shows better quality of peptide-MHC interactions.4Journal of Immunology Research influencing protein movements.The elastic network identifies pairs of atoms connected by springs.

TABLE 2 :
Interaction similarity scores of the selected HTL epitopes in CD70.

TABLE 3 :
Final selected CTL epitopes from CD70 protein for constructing multiepitope vaccine witht their corresponding MHC Class I alleles and immunogenic traits.
* Lower rates show better binding affinity.** Higher rate shows high degree of peptide antigenicity.Journal of Immunology Research

TABLE 4 :
Final selected HTL epitopes from CD70 protein for constructing multiepitope vaccine with their corresponding MHC Class II alleles and immunogenic traits.
*Lower rates show better binding affinity.**Higher rate shows high degree of peptide antigenicity.8Journal of Immunology Research

TABLE 5 :
MHC-I processing prediction and immunogenicity scores of CD70 CTL epitopes.
*Higher score shows better quality of TAP transport efficiency, * * higher score shows better quality of proteasomal cleavage, * * * higher rates show better quality of epitope identification, * * * higher score indicates a greater probability of eliciting an immune response.

TABLE 6 :
Physicochemical characteristics of the constructed vaccine.

TABLE 9 :
Protein-Protein docking results (lowest energy in the best model) between final vaccine constructs and TNF receptor.