Researchers conducted a comprehensive in silico study to evaluate the immunogenic potential of Heat Shock Protein 90 (Hsp90) from the parasite Toxoplasma gondii. Published in the journal Immunity, Inflammation and Disease, the study utilized bioinformatics tools to predict the protein’s structural features, physicochemical properties, and specific immune epitopes. The primary objective was to identify candidate antigens for subunit vaccine development against toxoplasmosis, a parasitic infection that currently lacks a licensed human vaccine. By focusing on Hsp90, a chaperone protein essential for parasite survival and development, the authors aimed to provide a rational basis for future experimental validation in preclinical models.
Toxoplasma gondii is an opportunistic parasite belonging to the phylum Apicomplexa, with felids serving as definitive hosts and humans as intermediate hosts. Infection typically occurs through the ingestion of undercooked meat containing tissue cysts or contaminated food and water. The parasite exists in two main forms: rapidly replicating tachyzoites that cause acute disease and slowly dividing bradyzoites that persist in tissue cysts. Current chemotherapeutic treatments primarily target tachyzoites but fail to eradicate tissue cysts, leading to a risk of reactivation in immunocompromised individuals. This limitation underscores the critical need for vaccine strategies that can provide durable protection against both acute and latent infection.
The study began by retrieving the amino acid sequence of T. gondii Hsp90 from the UniProtKB database, specifically using the accession number A0A125YVB7. The protein was characterized using Expasy ProtParam, which predicted a molecular weight of 81.93319 kDa and a sequence length of 708 amino acid residues. The calculated isoelectric point (pI) was 4.96, indicating a relatively acidic nature, while the grand average of hydropathicity (GRAVY) score of −0.626 suggested the protein is hydrophilic. These physicochemical properties are significant for potential recombinant expression and purification, as the hydrophilic nature may facilitate interaction in aqueous environments during vaccine production.
Post-translational modification (PTM) analysis revealed a complex modification pattern for Hsp90. The NetPhos 3.1 server identified 67 potential phosphorylation sites, distributed among 14 tyrosine, 23 threonine, and 30 serine residues. Additionally, the protein was predicted to have one acetylation site, four N-linked glycosylation sites, and one O-linked glycosylation site. The presence of multiple glycosylation sites suggests that heterologous expression of Hsp90 may require eukaryotic systems, such as yeast or mammalian cells, to ensure proper folding and functional maturation. Subcellular localization predictions using DeepLoc-2.1 indicated that Hsp90 is primarily located in the cytoplasm with a probability score of 0.7623, with secondary predictions for the nucleus and Golgi apparatus. The detection of a nuclear export signal implies that the protein may shuttle between the nucleus and cytoplasm, a feature common in regulatory and stress-responsive proteins.
Structural modeling was performed to understand the three-dimensional conformation of Hsp90. Secondary structure prediction using GOR IV, PSIPRED, and NetSurfP3 identified the protein as being composed of 50.85% alpha helix, 33.47% random coil, and 15.68% extended strand. The three-dimensional model was generated using the SWISS-MODEL server, selecting a template with 82.98% sequence identity. The model was subsequently refined using Galaxy Refine, which improved the structural quality as indicated by a MolProbity score of 1.464 and a Ramachandran favored region percentage of 98.6%. Validation using ProSA-web yielded a Z-score of −9.67 for the refined model, confirming its reliability for further epitope mapping. The absence of transmembrane domains was confirmed by DEEP-TMHMM, supporting the protein’s role as a soluble cytoplasmic effector.
The core of the study involved the prediction of B-cell and T-cell epitopes to assess the protein’s immunogenicity. Linear B-cell epitopes were identified using ABCpred, BcePred, BCPREDS, and SVMTriP. While many predicted epitopes showed high antigenicity scores, allergenicity analysis using AlgPred classified the majority as potential allergens. However, two specific epitopes, RVFIMDDCEDLIPEWL and GFSLDEPTQFAARIHR, were predicted to be nonallergenic while retaining favorable antigenicity and solubility. These two sequences were highlighted as priority candidates for future experimental validation due to their balanced safety and immunogenic profiles. Conformational B-cell epitopes were also identified using ElliPro, which detected 12 discontinuous epitopes with scores ranging from 0.503 to 0.994, further supporting the potential of Hsp90 to induce antibody-mediated immune responses.
T-cell epitope prediction focused on Major Histocompatibility Complex (MHC) binding and Cytotoxic T-Lymphocyte (CTL) activity. MHC-I and MHC-II specific epitopes were predicted for murine alleles, including H-2-Db, H-2-Kb, and H-2-IAd. The IEDB analysis revealed several peptides with low percentile ranks, indicating strong binding affinity. For instance, the peptide KGAERLFIRI showed a percentile rank of 1.1 and an immunogenicity score of 0.40581 for MHC-I binding. CTLpred analysis identified the top 10 predicted 9-mer CTL epitopes, with IVRGTRIIL ranking highest with an artificial neural network score of 0.94. These findings suggest that Hsp90 contains multiple regions capable of stimulating both humoral and cell-associated immunity, which is essential for comprehensive protection against intracellular parasites.
Antigenicity and solubility evaluations further supported the candidacy of Hsp90 for vaccine development. The protein yielded an antigenicity score of 0.757094 using ANTIGENpro and 0.5513 with VaxiJen, both exceeding the threshold of 0.5 for antigenic potential. Allergenicity assessment confirmed that the full-length Hsp90 protein is nonallergenic. Solubility prediction using SOLpro indicated a likelihood of 0.62436 for the protein to remain soluble after overexpression in E. coli. These combined metrics suggest that Hsp90 is a stable, nonallergenic, and highly antigenic protein, making it a robust candidate for subunit vaccine formulation.
The study acknowledges several limitations that must be considered when interpreting the results. The MHC binding predictions were performed exclusively using murine alleles, which serve as a primary screening step in immunoinformatics but do not fully represent human HLA diversity. The authors note that future studies must include human HLA Class I and II alleles to improve the translational relevance of the findings. Additionally, the study did not assess epitope conservation across different strains of T. gondii, which is crucial for ensuring broad-spectrum protection against the genetic variability of the parasite. The absence of in vivo validation means that the predicted epitopes remain theoretical candidates until confirmed through experimental immunization studies.
In conclusion, this computational study provides a detailed immunoinformatic profile of T. gondii Hsp90, highlighting its potential as a target for subunit vaccine development. The identification of specific nonallergenic B-cell epitopes and high-affinity MHC-binding T-cell epitopes offers a rational starting point for vaccine design. While the results are primarily applicable to mouse preclinical models, the comprehensive characterization of Hsp90’s structural and immunogenic features lays the groundwork for future research. The authors recommend that subsequent in vitro and in vivo validation be performed to assess the protein’s efficacy in protecting against toxoplasmosis, ultimately contributing to the development of a licensed human vaccine.
The selection of Hsp90 as a vaccine target is grounded in its biological function within the parasite. As a chaperone protein, Hsp90 manages essential processes related to cell growth and differentiation, including protein homeostasis, the refolding of misfolded proteins, and the regulation of nuclear hormone receptors and kinase signaling. Its expression and subcellular localization are developmentally regulated, particularly during the conversion between rapidly replicating tachyzoites and slowly dividing bradyzoites. This dynamic role makes it a promising immunological target, as it is consistently produced in large amounts and its expression rises in response to stressful conditions. The study aimed to characterize this protein using immunoinformatics to predict structural features, evaluate antigenicity, and identify cytotoxic T-lymphocyte, putative B-cell, and helper T-lymphocyte epitopes to inform rational vaccine design.
Bioinformatics tools allow researchers to detect possible B- and T-cell epitopes, thereby improving the accuracy of vaccine candidate development. These in silico methods, based on computer simulations rather than laboratory experiments, enable the prediction of antigenic proteins even without in vitro expression. The study utilized a multistep approach to identify linear B-cell epitopes using ABCpred, BcePred, BCPREDS, and SVMTriP. While several predicted B-cell epitopes showed high scores and favorable antigenicity, allergenicity analysis indicated that some ABCpred-derived peptides may have potential allergenic properties. Therefore, antigenicity alone was not considered sufficient for epitope prioritization. Instead, suitable vaccine candidates were required to combine high antigenicity, good solubility, and nonallergenic characteristics. Based on these criteria, the sequences RVFIMDDCEDLIPEWL and GFSLDEPTQFAARIHR were identified as promising B-cell epitope candidates for future experimental evaluation.
The interaction between peptides and MHC molecules is critical for T-cell antigen presentation and significantly affects epitope potential. Lower IC50 values indicate increased binding affinity, suggesting a better T-cell epitope, while higher values imply decreased affinity. For MHC-II, among all the forecasted epitopes, ANMERIMKAQALRDN was found to be the most antigenic peptide with a VaxiJen score of 0.7229. Furthermore, DKVEQVVVSNRITDS and SLTNDWEDPLAVKHF were identified as capable of inducing IFN-γ. Regarding IL-4 induction, all epitopes were strong inducers except for HDKVEQVVVSNRITD. For MHC-I alleles, the most immunogenic epitopes were KGAERLFIRI, SNKEIFLREL, and SEFISFPIEL. The Hsp90 protein itself was confirmed as an antigenic protein, with scores of 0.757094 and 0.5513 from ANTIGENpro and VaxiJen v2.0, respectively, both exceeding the threshold of 0.5.
A significant limitation of the study is that MHC binding predictions were performed exclusively using murine alleles. While murine models are widely used as a primary screening step in immunoinformatics-based vaccine design, they do not fully represent human HLA diversity. Future studies should include HLA Class I and II alleles to improve translational relevance and vaccine applicability in humans. Another limitation is the absence of epitope conservation analysis across different strains of T. gondii. Since genetic variability among parasite strains may influence epitope stability and vaccine efficacy, future studies should evaluate the conservation of predicted epitopes across multiple strains to ensure broad-spectrum protection. Additionally, human HLA allele coverage and population-wide immunogenicity were not assessed, which limits the direct extrapolation of results to human populations. Therefore, future work should incorporate HLA-based population coverage analysis to improve translational relevance.
Subject of Research: Computational Prediction of Immunogenic Epitopes and Protein Structures of Heat Shock Protein 90 in Toxoplasma gondii
Article Title: Computational Prediction of Immunogenic Epitopes and Protein Structures of Heat Shock Protein 90 in Toxoplasma gondii
Article References: Abedi, A., Ghaffari, A. D., Dehghanpour, S., & Fekrikia, M. (2026). Computational Prediction of Immunogenic Epitopes and Protein Structures of Heat Shock Protein 90 in Toxoplasma gondii. Immunity, Inflammation and Disease, 14(11), Article e70584. https://doi.org/10.1002/iid3.70584
Image Credits: AI Generated
DOI: 10.1002/iid3.70584
Keywords: Computational, Prediction, Immunogenic, Epitopes, Protein, Structures, Heat, Shock, Toxoplasma, gondii, scientific research
News Source: Kristina Jarvis. (November 9, 2026). Computational Study Maps Hsp90 Epitopes for Toxoplasma Vaccine Design. Scienmag.
