Predicting sequences and structures of MHC-binding peptides: a computational combinatorial approach

POTN: A Human Leukocyte Antigen-A2 Immunogenic Peptides Screening Model and Its Applications in Tumor Antigens Prediction

Whole genome/exome sequencing data for tumors are now abundant, and many tumor antigens, especially mutant antigens (neoantigens), have been identified for cancer immunotherapy. However, only a small fraction of the peptides from these antigens induce cytotoxic T cell responses. Therefore, efficient methods to identify these antigenic peptides are crucial. The current models of major histocompatibility complex (MHC) binding and antigenic prediction are still inaccurate. In this study, 360 9-mer peptides with verified immunological activity were selected to construct a prediction of tumor neoantigen (POTN) model, an immunogenic prediction model specifically for the human leukocyte antigen-A2 allele. Based on the physicochemical properties of amino acids, such as the residue propensity, hydrophobicity, and organic solvent/water, we found that the predictive capability of POTN is superior to that of the prediction programs SYPEITHI, IEDB, and NetMHCpan 4.0. We used POTN to screen peptides for the cancer-testis antigen located on the X chromosome, and we identified several peptides that may trigger immunogenicity. We synthesized and measured the binding affinity and immunogenicity of these peptides and found that the accuracy of POTN is higher than that of NetMHCpan 4.0. Identifying the properties related to the T cell response or immunogenicity paves the way to understanding the MHC/peptide/T cell receptor complex. In conclusion, POTN is an efficient prediction model for screening high-affinity immunogenic peptides from tumor antigens, and thus provides useful information for developing cancer immunotherapy.

A comprehensive review and performance evaluation of bioinformatics tools for HLA class I peptide-binding prediction

Human leukocyte antigen class I (HLA-I) molecules are encoded by major histocompatibility complex (MHC) class I loci in humans. The binding and interaction between HLA-I molecules and intracellular peptides derived from a variety of proteolytic mechanisms play a crucial role in subsequent T-cell recognition of target cells and the specificity of the immune response. In this context, tools that predict the likelihood for a peptide to bind to specific HLA class I allotypes are important for selecting the most promising antigenic targets for immunotherapy. In this article, we comprehensively review a variety of currently available tools for predicting the binding of peptides to a selection of HLA-I allomorphs. Specifically, we compare their calculation methods for the prediction score, employed algorithms, evaluation strategies and software functionalities. In addition, we have evaluated the prediction performance of the reviewed tools based on an independent validation data set, containing 21 101 experimentally verified ligands across 19 HLA-I allotypes. The benchmarking results show that MixMHCpred 2.0.1 achieves the best performance for predicting peptides binding to most of the HLA-I allomorphs studied, while NetMHCpan 4.0 and NetMHCcons 1.1 outperform the other machine learning-based and consensus-based tools, respectively. Importantly, it should be noted that a peptide predicted with a higher binding score for a specific HLA allotype does not necessarily imply it will be immunogenic. That said, peptide-binding predictors are still very useful in that they can help to significantly reduce the large number of epitope candidates that need to be experimentally verified. Several other factors, including susceptibility to proteasome cleavage, peptide transport into the endoplasmic reticulum and T-cell receptor repertoire, also contribute to the immunogenicity of peptide antigens, and some of them can be considered by some predictors. Therefore, integrating features derived from these additional factors together with HLA-binding properties by using machine-learning algorithms may increase the prediction accuracy of immunogenic peptides. As such, we anticipate that this review and benchmarking survey will assist researchers in selecting appropriate prediction tools that best suit their purposes and provide useful guidelines for the development of improved antigen predictors in the future.

Level of neo-epitope predecessor and mutation type determine T cell activation of MHC binding peptides

BACKGROUND

Targeting epitopes derived from neo-antigens (or "neo-epitopes") represents a promising immunotherapy approach with limited off-target effects. However, most peptides predicted using MHC binding prediction algorithms do not induce a CD8 + T cell response, and there is a crucial need to refine the predictions to readily identify the best antigens that could mediate T-cell responses. Such a response requires a high enough number of epitopes bound to the target MHC. This number is correlated with both the peptide-MHC binding affinity and the number of peptides reaching the ER. Beyond this, the response may be affected by the properties of the neo-epitope mutated residues.

METHODS

Herein, we analyzed several experimental datasets from cancer patients to elaborate better predictive algorithms for T-cell reactivity to neo-epitopes.

RESULTS

Indeed, potent classifiers for epitopes derived from neo-antigens in melanoma and other tumors can be developed based on biochemical properties of the mutated residue, the antigen expression level and the peptide processing stage. Among MHC binding peptides, the present classifiers can remove half of the peptides falsely predicted to activate T cells while maintaining the absolute majority of reactive peptides.

CONCLUSIONS

The classifier properties further highlight the contribution of the quantity of peptides reaching the ER and the mutation type to CD8 + T cell responses. These classifiers were then validated on neo-antigens obtained from other datasets, confirming the validity of our prediction. Algorithm Availability: http://peptibase.cs.biu.ac.il/Tcell_predictor/ or by request from the authors as a standalone code.

Computational Prediction of the Epitopes of HA1 Protein of Influenza Viruses to its Neutralizing Antibodies

In this work, we have used a new method to predict the epitopes of HA1 protein of influenza virus to several antibodies HC19, CR9114, BH151 and 4F5. While our results reproduced the binding epitopes of H3N2 or H5N1 for the neutralizing antibodies HC19, CR9114, and BH151 as revealed from the available crystal structures, additional epitopes for these antibodies were also suggested. Moreover, the predicted epitopes of H5N1 HA1 for the newly developed antibody 4F5 are located at the receptor binding domain, while previous study identified a region 76-WLLGNP-81 as the epitope. The possibility of antibody recognition of influenza virus via different mechanism by binding to different epitopes of an antigen is also discussed.

Computational identification of antibody epitopes of human papillomavirus 16 (HPV16) L1 proteins

Previously, we developed a method to predict epitopes on a protein recognized by specific antibodies. In this study, we have applied this method to identify the epitopes of the human papillomavirus 16 (HPV16) L1 capsomer that is bound by monoclonal antibodies U4, AE3 and AG7. Initially, the method was validated by the identification of epitopes of HPV16 L1 capsomer that bind to antibody U4. Our predicted epitopes were in agreement with the cryto-electron microscopy (cryto-EM) structure of the complex. The method was then used to predict the epitopes of HPV16 L1 binding of antibodies AE3 and AG7. Our calculations indicated that antibody AE3 binds to the HPV16 L1 capsomer at two different regions. Firstly, the region recognized by antibody U4 and secondly, the region recognized by antibody V5, which have been shown in the cryto-EM structure of the V5 and HPV16 L1 complex. In comparison, the antibody AG7 binds to the capsomer only at the epitopes bound by antibody U4. Therefore, antibody AE3 is predicted to have higher affinity than antibody AG7 and could be used for developing highly efficient anti-HPV monoclonal antibodies in the clinical treatment of HPV infections.

Design of peptide inhibitors of human papillomavirus 16 (HPV16) transcriptional regulator E1–E2 formation

Here, we have proposed a new scheme of the computational combinatorial design approach to identify potential inhibitor peptides. It consists of four steps: (i) using “multiple copy simultaneous search” (MCSS) procedure to locate specific functional groups on the protein surface; (ii) the peptide main chain is constructed based on the location of favored N-methylacetamide (NMA) groups; (iii) molecular dynamics simulations of the complex formed between the constructed peptides with the target protein in explicit water molecules are carried to select the peptides with strong binding to the protein and (iv) the sequences of the stable peptides selected from (iii) are aligned and the frequencies of the amino acids at each position of peptide are calculated. Sequence patterns of potential inhibitors are determined based on the frequency of amino acids at each position. It was applied to design peptide inhibitors that bind to the E2 protein of HPV16 so as to disrupt its transcriptional regulator of E1–E2 complex formation. The sequence pattern of these potential inhibitors is in agreement with known inhibitors obtained from phage display, and the MCSS calculations indicate that a hydrophobic pocket on HPV16 E2 plays a significant role in E1–E2 formation and inhibitor-E2 binding.

Computational Identification of Antibody Epitopes on the Dengue Virus NS1 Protein

We have previously described a method to predict antigenic epitopes on proteins recognized by specific antibodies. Here we have applied this method to identify epitopes on the NS1 proteins of the four Dengue virus serotypes (DENV1-4) that are bound by a small panel of monoclonal antibodies 1H7.4, 1G5.3 and Gus2. Several epitope regions were predicted for these antibodies and these were found to reflect the experimentally observed reactivities. The known binding epitopes on DENV2 for the antibodies 1H7.4 and 1G5.3 were identified, revealing the reasons for the serotype specificity of 1H7.4 and 1G5.3, and the non-selectivity of Gus2. As DENV NS1 is critical for virus replication and a key vaccine candidate, epitope prediction will be valuable in designing appropriate vaccine control strategies. The ability to predict potential epitopes by computational methods significantly reduces the amount of experimental work required to screen peptide libraries for epitope mapping.

Computer-aided vaccine designing approach against fish pathogens Edwardsiella tarda and Flavobacterium columnare using bioinformatics softwares

Edwardsiella tarda and Flavobacterium columnare are two important intracellular pathogenic bacteria that cause the infectious diseases edwardsiellosis and columnaris in wild and cultured fish. Prediction of major histocompatibility complex (MHC) binding is an important issue in T-cell epitope prediction. In a healthy immune system, the T-cells must recognize epitopes and induce the immune response. In this study, T-cell epitopes were predicted by using in silico immunoinformatics approach with the help of bioinformatics tools that are less expensive and are not time consuming. Such identification of binding interaction between peptides and MHC alleles aids in the discovery of new peptide vaccines. We have reported the potential peptides chosen from the outer membrane proteins (OMPs) of E. tarda and F. columnare, which interact well with MHC class I alleles. OMPs from E. tarda and F. columnare were selected and analyzed based on their antigenic and immunogenic properties. The OMPs of the genes TolC and FCOL_04620, respectively, from E. tarda and F. columnare were taken for study. Finally, two epitopes from the OMP of E. tarda exhibited excellent protein-peptide interaction when docked with MHC class I alleles. Five epitopes from the OMP of F. columnare had good protein-peptide interaction when docked with MHC class I alleles. Further in vitro studies can aid in the development of potential peptide vaccines using the predicted peptides.

Antibody recognition of Shiga toxins (Stxs): computational identification of the epitopes of Stx2 subunit A to the antibodies 11E10 and S2C4

We have recently developed a new method to predict the epitopes of the antigens that are recognized by a specific antibody. In this work, we applied the method to identify the epitopes of the Shiga toxin (Stx2 subunit A) that were bound by two specific antibodies 11E10 and S2C4. The predicted epitopes of Stx2 binding to the antibody 11E10 resembles the recognition surface constructed by the regions of Stx2 identified experimentally. For the S2C4, our results indicate that the antibody recognizes the Stx2 at two different regions on the protein surface. The first region (residues 246-254: ARSVRAVNE) is similar to the recognition region of the 11E10, while the second region is formed by two epitopes. The second region is particularly significant because it includes the amino acid sequence region that is diverse between Stx2 and other Stx (residues 176-188: QREFRQALSETAPV). This new recognition region is believed to play an important role in the experimentally observed selectivity of S2C4 to the Stx2.

Computational identification of epitopes in the glycoproteins of novel bunyavirus (SFTS virus) recognized by a human monoclonal antibody (MAb 4-5)

In this work, we have developed a new approach to predict the epitopes of antigens that are recognized by a specific antibody. Our method is based on the "multiple copy simultaneous search" (MCSS) approach which identifies optimal locations of small chemical functional groups on the surfaces of the antibody, and identifying sequence patterns of peptides that can bind to the surface of the antibody. The identified sequence patterns are then used to search the amino-acid sequence of the antigen protein. The approach was validated by reproducing the binding epitope of HIV gp120 envelop glycoprotein for the human neutralizing antibody as revealed in the available crystal structure. Our method was then applied to predict the epitopes of two glycoproteins of a newly discovered bunyavirus recognized by an antibody named MAb 4-5. These predicted epitopes can be verified by experimental methods. We also discuss the involvement of different amino acids in the antigen-antibody recognition based on the distributions of MCSS minima of different functional groups.

Quantitative prediction of peptide binding to HLA-DP1 protein

The exogenous proteins are processed by the host antigen-processing cells. Peptidic fragments of them are presented on the cell surface bound to the major hystocompatibility complex (MHC) molecules class II and recognized by the CD4+ T lymphocytes. The MHC binding is considered as the crucial prerequisite for T-cell recognition. Only peptides able to form stable complexes with the MHC proteins are recognized by the T-cells. These peptides are known as T-cell epitopes. All T-cell epitopes are MHC binders, but not all MHC binders are T-cell epitopes. The T-cell epitope prediction is one of the main priorities of immunoinformatics. In the present study, three chemometric techniques are combined to derive a model for in silico prediction of peptide binding to the human MHC class II protein HLA-DP1. The structures of a set of known peptide binders are described by amino acid z-descriptors. Data are processed by an iterative self-consisted algorithm using the method of partial least squares, and a quantitative matrix (QM) for peptide binding prediction to HLA-DP1 is derived. The QM is validated by two sets of proteins and showed an average accuracy of 86 percent.

T-cell epitope vaccine design by immunoinformatics

Vaccination is generally considered to be the most effective method of preventing infectious diseases. All vaccinations work by presenting a foreign antigen to the immune system in order to evoke an immune response. The active agent of a vaccine may be intact but inactivated ('attenuated') forms of the causative pathogens (bacteria or viruses), or purified components of the pathogen that have been found to be highly immunogenic. The increased understanding of antigen recognition at molecular level has resulted in the development of rationally designed peptide vaccines. The concept of peptide vaccines is based on identification and chemical synthesis of B-cell and T-cell epitopes which are immunodominant and can induce specific immune responses. The accelerating growth of bioinformatics techniques and applications along with the substantial amount of experimental data has given rise to a new field, called immunoinformatics. Immunoinformatics is a branch of bioinformatics dealing with in silico analysis and modelling of immunological data and problems. Different sequence- and structure-based immunoinformatics methods are reviewed in the paper.

An introduction to epitope prediction methods and software

In this paper, current prediction methods and algorithms for both T- and B cell epitopes are reviewed, and a comprehensive summary of epitope prediction software and databases currently available online is also provided. This review can offer researchers in this field a sense of direction and insights for future work. However, our main purpose is to introduce clinical and basic biomedical researchers who are not familiar with these biological analysis tools and databases to these online resources and to provide guidance on how to use them effectively.

HLA-peptide binding prediction using structural and modeling principles

Short peptides binding to specific human leukocyte antigen (HLA) alleles elicit immune response. These candidate peptides have potential utility in peptide vaccine design and development. The binding of peptides to allele-specific HLA molecule is estimated using competitive binding assay and biochemical binding constants. Application of this method for proteome-wide screening in parasites, viruses, and virulent bacterial strains is laborious and expensive. However, short listing of candidate peptides using prediction approaches have been realized lately. Prediction of peptide binding to HLA alleles using structural and modeling principles has gained momentum in recent years. Here, we discuss the current status of such prediction.