Integrating In Silico and In Vitro Analysis of Peptide Binding Affinity to HLA-Cw*0102: A Bioinformatic Approach to the Prediction of New Epitopes

Background

Predictive models of peptide-Major Histocompatibility Complex (MHC) binding affinity are important components of modern computational immunovaccinology. Here, we describe the development and deployment of a reliable peptide-binding prediction method for a previously poorly-characterized human MHC class I allele, HLA-Cw*0102.

Methodology/Findings

Using an in-house, flow cytometry-based MHC stabilization assay we generated novel peptide binding data, from which we derived a precise two-dimensional quantitative structure-activity relationship (2D-QSAR) binding model. This allowed us to explore the peptide specificity of HLA-Cw*0102 molecule in detail. We used this model to design peptides optimized for HLA-Cw*0102-binding. Experimental analysis showed these peptides to have high binding affinities for the HLA-Cw*0102 molecule. As a functional validation of our approach, we also predicted HLA-Cw*0102-binding peptides within the HIV-1 genome, identifying a set of potent binding peptides. The most affine of these binding peptides was subsequently determined to be an epitope recognized in a subset of HLA-Cw*0102-positive individuals chronically infected with HIV-1.

Conclusions/Significance

A functionally-validated in silico-in vitro approach to the reliable and efficient prediction of peptide binding to a previously uncharacterized human MHC allele HLA-Cw*0102 was developed. This technique is generally applicable to all T cell epitope identification problems in immunology and vaccinology.     

Peptide length significantly influences in vitro affinity for MHC class II molecules

Background

Class II Major Histocompatibility Complex (MHC) molecules have an open-ended binding groove which can accommodate peptides of varying lengths. Several studies have demonstrated that peptide flanking residues (PFRs) which lie outside the core binding groove can influence peptide binding and T cell recognition. By using data from the AntiJen database we were able to characterise systematically the influence of PFRs on peptide affinity for MHC class II molecules.

Results

By analysing 1279 peptide elongation events covering 19 distinct HLA alleles it was observed that, in general, peptide elongation resulted in increased MHC class II molecule affinity. It was also possible to determine an optimal peptide length for MHC class II affinity of approximately 18–20 amino acids; elongation of peptides beyond this length resulted in a null or negative effect on affinity.

Conclusion

The observed relationship between peptide length and MHC class II affinity has significant implications for the design of vaccines and the study of the epitopic basis of immunological disease.     

Molecular recognition of H3/H4 histone tails by the tudor domains of JMJD2A: a comparative molecular dynamics simulations study

Background

Histone demethylase, JMJD2A, specifically recognizes and binds to methylated lysine residues at histone H3 and H4 tails (especially trimethylated H3K4 (H3K4me3), trimethylated H3K9 (H3K9me3) and di,trimethylated H4K20 (H4K20me2, H4K20me3)) via its tandem tudor domains. Crystal structures of JMJD2A-tudor binding to H3K4me3 and H4K20me3 peptides are available whereas the others are not. Complete picture of the recognition of the four histone peptides by the tandem tudor domains yet remains to be clarified.

Methodology/Principal Findings

We report a detailed molecular dynamics simulation and binding energy analysis of the recognition of JMJD2A-tudor with four different histone tails. 25 ns fully unrestrained molecular dynamics simulations are carried out for each of the bound and free structures. We investigate the important hydrogen bonds and electrostatic interactions between the tudor domains and the peptide molecules and identify the critical residues that stabilize the complexes. Our binding free energy calculations show that H4K20me2 and H3K9me3 peptides have the highest and lowest affinity to JMJD2A-tudor, respectively. We also show that H4K20me2 peptide adopts the same binding mode with H4K20me3 peptide, and H3K9me3 peptide adopts the same binding mode with H3K4me3 peptide. Decomposition of the enthalpic and the entropic contributions to the binding free energies indicate that the recognition of the histone peptides is mainly driven by favourable van der Waals interactions. Residue decomposition of the binding free energies with backbone and side chain contributions as well as their energetic constituents identify the hotspots in the binding interface of the structures.

Conclusion

Energetic investigations of the four complexes suggest that many of the residues involved in the interactions are common. However, we found two receptor residues that were related to selective binding of the H3 and H4 ligands. Modifications or mutations on one of these residues can selectively alter the recognition of the H3 tails or the H4 tails.     

Peptide-MHC Cellular Microarray with Innovative Data Analysis System for Simultaneously Detecting Multiple CD4 T-Cell Responses

Background

Peptide:MHC cellular microarrays have been proposed to simultaneously characterize multiple Ag-specific populations of T cells. The practice of studying immune responses to complicated pathogens with this tool demands extensive knowledge of T cell epitopes and the availability of peptide:MHC complexes for array fabrication as well as a specialized data analysis approach for result interpretation.

Methodology/Principal Findings

We co-immobilized peptide:DR0401 complexes, anti-CD28, anti-CD11a and cytokine capture antibodies on the surface of chamber slides to generate a functional array that was able to detect rare Ag-specific T cell populations from previously primed in vitro T cell cultures. A novel statistical methodology was also developed to facilitate batch processing of raw array-like data into standardized endpoint scores, which linearly correlated with total Ag-specific T cell inputs. Applying these methods to analyze Influenza A viral antigen-specific T cell responses, we not only revealed the most prominent viral epitopes, but also demonstrated the heterogeneity of anti-viral cellular responses in healthy individuals. Applying these methods to examine the insulin producing beta-cell autoantigen specific T cell responses, we observed little difference between autoimmune diabetic patients and healthy individuals, suggesting a more subtle association between diabetes status and peripheral autoreactive T cells.

Conclusions/Significance

The data analysis system is reliable for T cell specificity and functional testing. Peptide:MHC cellular microarrays can be used to obtain multi-parametric results using limited blood samples in a variety of translational settings.     

HLA-DM Mediates Epitope Selection by a “Compare-Exchange” Mechanism when a Potential Peptide Pool Is Available

Background

HLA-DM (DM) mediates exchange of peptides bound to MHC class II (MHCII) during the epitope selection process. Although DM has been shown to have two activities, peptide release and MHC class II refolding, a clear characterization of the mechanism by which DM facilitates peptide exchange has remained elusive.

Methodology/Principal Findings

We have previously demonstrated that peptide binding to and dissociation from MHCII in the absence of DM are cooperative processes, likely related to conformational changes in the peptide-MHCII complex. Here we show that DM promotes peptide release by a non-cooperative process, whereas it enhances cooperative folding of the exchange peptide. Through electron paramagnetic resonance (EPR) and fluorescence polarization (FP) we show that DM releases prebound peptide very poorly in the absence of a candidate peptide for the exchange process. The affinity and concentration of the candidate peptide are also important for the release of the prebound peptide. Increased fluorescence energy transfer between the prebound and exchange peptides in the presence of DM is evidence for a tetramolecular complex which resolves in favor of the peptide that has superior folding properties.

Conclusion/Significance

This study shows that both the peptide releasing activity on loaded MHCII and the facilitating of MHCII binding by a candidate exchange peptide are integral to DM mediated epitope selection. The exchange process is initiated only in the presence of candidate peptides, avoiding possible release of a prebound peptide and loss of a potential epitope. In a tetramolecular transitional complex, the candidate peptides are checked for their ability to replace the pre-bound peptide with a geometry that allows the rebinding of the original peptide. Thus, DM promotes a “compare-exchange” sorting algorithm on an available peptide pool. Such a “third party”-mediated mechanism may be generally applicable for diverse ligand recognition in other biological systems.     

Targeting of Embryonic Stem Cells by Peptide-Conjugated Quantum Dots

Background

Targeting stem cells holds great potential for studying the embryonic stem cell and development of stem cell-based regenerative medicine. Previous studies demonstrated that nanoparticles can serve as a robust platform for gene delivery, non-invasive cell imaging, and manipulation of stem cell differentiation. However specific targeting of embryonic stem cells by peptide-linked nanoparticles has not been reported.

Methodology/Principal Findings

Here, we developed a method for screening peptides that specifically recognize rhesus macaque embryonic stem cells by phage display and used the peptides to facilitate quantum dot targeting of embryonic stem cells. Through a phage display screen, we found phages that displayed an APWHLSSQYSRT peptide showed high affinity and specificity to undifferentiated primate embryonic stem cells in an enzyme-linked immunoabsorbent assay. These results were subsequently confirmed by immunofluoresence microscopy. Additionally, this binding could be completed by the chemically synthesized APWHLSSQYSRT peptide, indicating that the binding capability was specific and conferred by the peptide sequence. Through the ligation of the peptide to CdSe-ZnS core-shell nanocrystals, we were able to, for the first time, target embryonic stem cells through peptide-conjugated quantum dots.

Conclusions/Significance

These data demonstrate that our established method of screening for embryonic stem cell specific binding peptides by phage display is feasible. Moreover, the peptide-conjugated quantum dots may be applicable for embryonic stem cell study and utilization.     

Dendritic cells reveal a broad range of MHC class I epitopes for HIV-1 in persons with suppressed viral load on antiretroviral therapy

Background

HIV-1 remains sequestered during antiretroviral therapy (ART) and can resume high-level replication upon cessation of ART or development of drug resistance. Reactivity of memory CD8⁺ T lymphocytes to HIV-1 could potentially inhibit this residual viral replication, but is largely muted by ART in relation to suppression of viral antigen burden. Dendritic cells (DC) are important for MHC class I processing and presentation of peptide epitopes to memory CD8⁺ T cells, and could potentially be targeted to activate memory CD8⁺ T cells to a broad array of HIV-1 epitopes during ART.

Principal Findings

We show for the first time that HIV-1 peptide-loaded, CD40L-matured DC from HIV-1 infected persons on ART induce IFN gamma production by CD8⁺ T cells specific for a much broader range and magnitude of Gag and Nef epitopes than do peptides without DC. The DC also reveal novel, MHC class I restricted, Gag and Nef epitopes that are able to induce polyfunctional T cells producing various combinations of IFN gamma, interleukin 2, tumor necrosis factor alpha, macrophage inhibitory protein 1 beta and the cytotoxic de-granulation molecule CD107a.

Significance

There is an underlying, broad antigenic spectrum of anti-HIV-1, memory CD8⁺ T cell reactivity in persons on ART that is revealed by DC. This supports the use of DC-based immunotherapy for HIV-1 infection.     

One-pot, mix-and-read peptide-MHC tetramers

Background

Cytotoxic T Lymphocytes (CTL) recognize complexes of peptide ligands and Major Histocompatibility Complex (MHC) class I molecules presented at the surface of Antigen Presenting Cells (APC). Detection and isolation of CTL''s are of importance for research on CTL immunity, and development of vaccines and adoptive immune therapy. Peptide-MHC tetramers have become important reagents for detection and enumeration of specific CTL''s. Conventional peptide-MHC-tetramer production involves recombinant MHC production, in vitro refolding, biotinylation and tetramerization; each step followed by various biochemical steps such as chromatographic purification, concentration etc. Such cumbersome production protocols have limited dissemination and restricted availability of peptide-MHC tetramers effectively precluding large-scale screening strategies involving many different peptide-MHC tetramers.

Methodology/Principal Findings

We have developed an approach whereby any given tetramer specificity can be produced within 2 days with very limited effort and hands-on time. The strategy is based on the isolation of correctly oxidized, in vivo biotinylated recombinant MHC I heavy chain (HC). Such biotinylated MHC I HC molecules can be refolded in vitro, tetramerized with streptavidin, and used for specific T cell staining-all in a one-pot reaction without any intervening purification steps.

Conclusions/Significance

We have developed an efficient “one-pot, mix-and-read” strategy for peptide-MHC tetramer generation, and demonstrated specific T cell straining comparable to a commercially available MHC-tetramer. Here, seven peptide-MHC tetramers representing four different human MHC (HLA) class I proteins have been generated. The technique should be readily extendable to any binding peptide and pre-biotinylated MHC (at this time we have over 40 different pre-biotinylated HLA proteins). It is simple, robust, and versatile technique with a very broad application potential as it can be adapted both to small- and large-scale production of one or many different peptide-MHC tetramers for T cell isolation, or epitope screening.     

Identification of Prostate-Specific G-Protein Coupled Receptor as a Tumor Antigen Recognized by CD8+ T Cells for Cancer Immunotherapy

Background

Prostate cancer is the most common cancer among elderly men in the US, and immunotherapy has been shown to be a promising strategy to treat patients with metastatic castration-resistant prostate cancer. Efforts to identify novel prostate specific tumor antigens will facilitate the development of effective cancer vaccines against prostate cancer. Prostate-specific G-protein coupled receptor (PSGR) is a novel antigen that has been shown to be specifically over-expressed in human prostate cancer tissues. In this study, we describe the identification of PSGR-derived peptide epitopes recognized by CD8⁺ T cells in an HLA-A2 dependent manner.

Methodology/Principal Findings

Twenty-one PSGR-derived peptides were predicted by an immuno-informatics approach based on the HLA-A2 binding motif. These peptides were examined for their ability to induce peptide-specific T cell responses in peripheral blood mononuclear cells (PBMCs) obtained from either HLA-A2⁺ healthy donors or HLA-A2⁺ prostate cancer patients. The recognition of HLA-A2 positive and PSGR expressing LNCaP cells was also tested. Among the 21 PSGR-derived peptides, three peptides, PSGR3, PSGR4 and PSGR14 frequently induced peptide-specific T cell responses in PBMCs from both healthy donors and prostate cancer patients. Importantly, these peptide-specific T cells recognized and killed LNCaP prostate cancer cells in an HLA class I-restricted manner.

Conclusions/Significance

We have identified three novel HLA-A2-restricted PSGR-derived peptides recognized by CD8⁺ T cells, which, in turn, recognize HLA-A2⁺ and PSGR⁺ tumor cells. The PSGR-derived peptides identified may be used as diagnostic markers as well as immune targets for development of anticancer vaccines.     

Emulsified Phosphatidylserine,Simple and Effective Peptide Carrier for Induction of Potent Epitope-Specific T Cell Responses

Background

To induce potent epitope-specific T cell immunity by a peptide-based vaccine, epitope peptides must be delivered efficiently to antigen-presenting cells (APCs) in vivo. Therefore, selecting an appropriate peptide carrier is crucial for the development of an effective peptide vaccine. In this study, we explored new peptide carriers which show enhancement in cytotoxic T lymphocyte (CTL) induction capability.

Methodology/Principal Findings

Data from an epitope-specific in vivo CTL assay revealed that phosphatidylserine (PS) has a potent adjuvant effect among candidate materials tested. Further analyses showed that PS-conjugated antigens were preferentially and efficiently captured by professional APCs, in particular, by CD11c⁺CD11b⁺MHCII⁺ conventional dendritic cells (cDCs) compared to multilamellar liposome-conjugates or unconjugated antigens. In addition, PS demonstrated the stimulatory capacity of peptide-specific helper T cells in vivo.

Conclusions/Significance

This work indicates that PS is the easily preparable efficient carrier with a simple structure that delivers antigen to professional APCs effectively and induce both helper and cytotoxic T cell responses in vivo. Therefore, PS is a promising novel adjuvant for T cell-inducing peptide vaccines.     

Comparative analysis of collagen type II-specific immune responses during development of collagen-induced arthritis in two B10 mouse strains

Introduction

Immune responses against collagen type II (CII) are crucial for the development of collagen-induced arthritis (CIA). The aim of the present study was to evaluate and compare the CII-directed T cell and antibody specificity at different time points in the course of CIA using two mouse strains on the B10 genetic background - B10.Q, expressing A^q MHC class II molecules, and B10.DR4.Ncf1^*/*, expressing human rheumatoid arthritis-associated MHC II DR4 molecules (DRA*0101/DRB*0401).

Methods

B10.Q and B10.DR4.Ncf1^*/* mice were immunized with CII emulsified in adjuvant and development of CIA was assessed. T cells from draining lymph nodes were restimulated in vitro with CII peptides and interferon-gamma (IFN-γ) levels in culture supernatants were evaluated by ELISA. CII-specific antibody levels in serum samples were measured by ELISA.

Results

At four different CIA time points we analyzed T cell specificity to the immunodominant CII epitope 259-273 (CII259-273) and several posttranslationally modified forms of CII259-273 as well as antibody responses to three B cell immunodominant epitopes on CII (C1, U1, J1). Our data show that CII-specific T and B cell responses increase dramatically after disease onset in both strains and are sustained during the disease course. Concerning anti-CII antibody fine specificity, during all investigated stages of CIA the B10.Q mice responded predominantly to the C1 epitope, whereas the B10.DR4.Ncf1^*/* mice also recognized the U1 epitope. In the established disease phase, T cell reactivity toward the galactosylated CII259-273 peptide was similar between the DR4- and the A^q-expressing strains whereas the response to the non-modified CII peptide was dramatically enhanced in the DR4 mice compared with the B10.Q. In addition, we show that the difference in the transgenic DR4-restricted T cell specificity to CII259-273 is not dependent on the degree of glycosylation of the collagen used for immunization.

Conclusions

The present study provides important evaluation of CII-specific immune responses at different phases during CIA development as well as a comparative analysis between two CIA mouse models. We indicate significant differences in CII T cell and antibody specificities between the two strains and highlight a need for improved humanized B10.DR4 mouse model for rheumatoid arthritis.     

SdPI, the first functionally characterized Kunitz-type trypsin inhibitor from scorpion venom

Background

Kunitz-type venom peptides have been isolated from a wide variety of venomous animals. They usually have protease inhibitory activity or potassium channel blocking activity, which by virtue of the effects on predator animals are essential for the survival of venomous animals. However, no Kunitz-type peptides from scorpion venom have been functionally characterized.

Principal Findings

A new Kunitz-type venom peptide gene precursor, SdPI, was cloned and characterized from a venom gland cDNA library of the scorpion Lychas mucronatus. It codes for a signal peptide of 21 residues and a mature peptide of 59 residues. The mature SdPI peptide possesses a unique cysteine framework reticulated by three disulfide bridges, different from all reported Kunitz-type proteins. The recombinant SdPI peptide was functionally expressed. It showed trypsin inhibitory activity with high potency (K_i = 1.6×10⁻⁷ M) and thermostability.

Conclusions

The results illustrated that SdPI is a potent and stable serine protease inhibitor. Further mutagenesis and molecular dynamics simulation revealed that SdPI possesses a serine protease inhibitory active site similar to other Kunitz-type venom peptides. To our knowledge, SdPI is the first functionally characterized Kunitz-type trypsin inhibitor derived from scorpion venom, and it represents a new class of Kunitz-type venom peptides.     

Identification of Special AT-Rich Sequence Binding Protein 1 as a Novel Tumor Antigen Recognized by CD8+ T Cells: Implication for Cancer Immunotherapy

Background

A large number of human tumor-associated antigens that are recognized by CD8⁺ T cells in a human leukocyte antigen class I (HLA-I)-restricted fashion have been identified. Special AT-rich sequence binding protein 1 (SATB1) is highly expressed in many types of human cancers as part of their neoplastic phenotype, and up-regulation of SATB1 expression is essential for tumor survival and metastasis, thus this protein may serve as a rational target for cancer vaccines.

Methodology/Principal Findings

Twelve SATB1-derived peptides were predicted by an immuno-informatics approach based on the HLA-A*02 binding motif. These peptides were examined for their ability to induce peptide-specific T cell responses in peripheral blood mononuclear cells (PBMCs) obtained from HLA-A*02⁺ healthy donors and/or HLA-A*02⁺ cancer patients. The recognition of HLA-A*02⁺ SATB1-expressing cancer cells was also tested. Among the twelve SATB1-derived peptides, SATB1_565–574 frequently induced peptide-specific T cell responses in PBMCs from both healthy donors and cancer patients. Importantly, SATB1_565–574-specific T cells recognized and killed HLA-A*02⁺ SATB1⁺ cancer cells in an HLA-I-restricted manner.

Conclusions/Significance

We have identified a novel HLA-A*02-restricted SATB1-derived peptide epitope recognized by CD8⁺ T cells, which, in turn, recognizes and kills HLA-A*02⁺ SATB1⁺ tumor cells. The SATB1-derived epitope identified may be used as a diagnostic marker as well as an immune target for development of cancer vaccines.     

VitAL: Viterbi Algorithm for de novo Peptide Design

Background

Drug design against proteins to cure various diseases has been studied for several years. Numerous design techniques were discovered for small organic molecules for specific protein targets. The specificity, toxicity and selectivity of small molecules are hard problems to solve. The use of peptide drugs enables a partial solution to the toxicity problem. There has been a wide interest in peptide design, but the design techniques of a specific and selective peptide inhibitor against a protein target have not yet been established.

Methodology/Principal Findings

A novel de novo peptide design approach is developed to block activities of disease related protein targets. No prior training, based on known peptides, is necessary. The method sequentially generates the peptide by docking its residues pair by pair along a chosen path on a protein. The binding site on the protein is determined via the coarse grained Gaussian Network Model. A binding path is determined. The best fitting peptide is constructed by generating all possible peptide pairs at each point along the path and determining the binding energies between these pairs and the specific location on the protein using AutoDock. The Markov based partition function for all possible choices of the peptides along the path is generated by a matrix multiplication scheme. The best fitting peptide for the given surface is obtained by a Hidden Markov model using Viterbi decoding. The suitability of the conformations of the peptides that result upon binding on the surface are included in the algorithm by considering the intrinsic Ramachandran potentials.

Conclusions/Significance

The model is tested on known protein-peptide inhibitor complexes. The present algorithm predicts peptides that have better binding energies than those of the existing ones. Finally, a heptapeptide is designed for a protein that has excellent binding affinity according to AutoDock results.     

Enhancement of Tumour-Specific Immune Responses In Vivo by ‘MHC Loading-Enhancer’ (MLE)

Background

Class II MHC molecules (MHC II) are cell surface receptors displaying short protein fragments for the surveillance by CD4+ T cells. Antigens therefore have to be loaded onto this receptor in order to induce productive immune responses. On the cell surface, most MHC II molecules are either occupied by ligands or their binding cleft has been blocked by the acquisition of a non-receptive state. Direct loading with antigens, as required during peptide vaccinations, is therefore hindered.

Principal Findings

Here we show, that the in vivo response of CD4+ T cells can be improved, when the antigens are administered together with ‘MHC-loading enhancer’ (MLE). MLE are small catalytic compounds able to open up the MHC binding site by triggering ligand-release and stabilizing the receptive state. Their enhancing effect on the immune response was demonstrated here with an antigen from the influenza virus and tumour associated antigens (TAA) derived from the NY-ESO-1 protein. The application of these antigens in combination with adamantane ethanol (AdEtOH), an MLE compound active on human HLA-DR molecules, significantly increased the frequency of antigen-specific CD4+ T cells in mice transgenic for the human MHC II molecule. Notably, the effect was evident only with the MLE-susceptible HLA-DR molecule and not with murine MHC II molecules non-susceptible for the catalytic effect of the MLE.

Conclusion

MLE can specifically increase the potency of a vaccine by facilitating the efficient transfer of the antigen onto the MHC molecule. They may therefore open a new way to improve vaccination efficacy and tumour-immunotherapy.     

Different Strategies Adopted by Kb and Ld to Generate T Cell Specificity Directed against Their Respective Bound Peptides

Mouse T cell clone 2C recognizes two different major histocompatibility (MHC) ligands, the self MHC K^b and the allogeneic MHC L^d. Two distinct peptides, SIY (SIYRYYGL) and QL9 (QLSPFPFDL), act as strong and specific agonists when bound to K^b and L^d, respectively. To explore further the mechanisms involved in peptide potency and specificity, here we examined a collection of single amino acid peptide variants of SIY and QL9 for 1) T cell activity, 2) binding to their respective MHC, and 3) binding to the 2C T cell receptor (TCR) and high affinity TCR mutants. Characterization of SIY binding to MHC K^b revealed significant effects of three SIY residues that were clearly embedded within the K^b molecule. In contrast, QL9 binding to MHC L^d was influenced by the majority of peptide side chains, distributed across the entire length of the peptide. Binding of the SIY-K^b complex to the TCR involved three SIY residues that were pointed toward the TCR, whereas again the majority of QL9 residues influenced binding of TCRs, and thus the QL9 residues had impacts on both L^d and TCR binding. In general, the magnitude of T cell activity mediated by a peptide variant was influenced more by peptide binding to MHC than by binding the TCR, especially for higher affinity TCRs. Findings with both systems, but QL9-L^d in particular, suggest that many single-residue substitutions, introduced into peptides to improve their binding to MHC and thus their vaccine potential, could impair T cell reactivity due to their dual impact on TCR binding.     

Dataset size and composition impact the reliability of performance benchmarks for peptide-MHC binding predictions

Background

It is important to accurately determine the performance of peptide:MHC binding predictions, as this enables users to compare and choose between different prediction methods and provides estimates of the expected error rate. Two common approaches to determine prediction performance are cross-validation, in which all available data are iteratively split into training and testing data, and the use of blind sets generated separately from the data used to construct the predictive method. In the present study, we have compared cross-validated prediction performances generated on our last benchmark dataset from 2009 with prediction performances generated on data subsequently added to the Immune Epitope Database (IEDB) which served as a blind set.

Results

We found that cross-validated performances systematically overestimated performance on the blind set. This was found not to be due to the presence of similar peptides in the cross-validation dataset. Rather, we found that small size and low sequence/affinity diversity of either training or blind datasets were associated with large differences in cross-validated vs. blind prediction performances. We use these findings to derive quantitative rules of how large and diverse datasets need to be to provide generalizable performance estimates.

Conclusion

It has long been known that cross-validated prediction performance estimates often overestimate performance on independently generated blind set data. We here identify and quantify the specific factors contributing to this effect for MHC-I binding predictions. An increasing number of peptides for which MHC binding affinities are measured experimentally have been selected based on binding predictions and thus are less diverse than historic datasets sampling the entire sequence and affinity space, making them more difficult benchmark data sets. This has to be taken into account when comparing performance metrics between different benchmarks, and when deriving error estimates for predictions based on benchmark performance.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2105-15-241) contains supplementary material, which is available to authorized users.     

Characterization of the specific CD4+ T cell response against the F protein during chronic hepatitis C virus infection

Background

The hepatitis C virus (HCV) Alternate Reading Frame Protein (ARFP or F protein) presents a double-frame shift product of the HCV core gene. We and others have previously reported that the specific antibodies against the F protein could be raised in the sera of HCV chronically infected patients. However, the specific CD4⁺ T cell responses against the F protein during HCV infection and the pathological implications remained unclear. In the current study, we screened the MHC class II-presenting epitopes of the F protein through HLA-transgenic mouse models and eventually validated the specific CD4⁺ T cell responses in HCV chronically infected patients.

Methodology

DNA vaccination in HLA-DR1 and-DP4 transgenic mouse models, proliferation assay to test the F protein specific T cell response, genotyping of Chronic HCV patients and testing the F-peptide stimulated T cell response in the peripheral blood mononuclear cell (PBMC) by in vitro expansion and interferon (IFN)- γ intracellular staining.

Principal Findings

At least three peptides within HCV F protein were identified as HLA-DR or HLA-DP4 presenting epitopes by the proliferation assays in mouse models. Further study with human PBMCs evidenced the specific CD4⁺ T cell responses against HCV F protein as well in patients chronically infected with HCV.

Conclusion

The current study provided the evidence for the first time that HCV F protein could elicit specific CD4⁺ T cell response, which may provide an insight into the immunopathogenesis during HCV chronic infection.     

In Vitro Priming Recapitulates In Vivo HIV-1 Specific T Cell Responses,Revealing Rapid Loss of Virus Reactive CD4+ T Cells in Acute HIV-1 Infection

Background

The requirements for priming of HIV-specific T cell responses initially seen in infected individuals remain to be defined. Activation of T cell responses in lymph nodes requires cell-cell contact between T cells and DCs, which can give concurrent activation of T cells and HIV transmission.

Methodology

The study aim was to establish whether DCs pulsed with HIV-1 could prime HIV-specific T cell responses and to characterize these responses. Both infectious and aldrithiol-2 inactivated noninfectious HIV-1 were compared to establish efficiencies in priming and the type of responses elicited.

Findings

Our findings show that both infectious and inactivated HIV-1 pulsed DCs can prime HIV-specific responses from naïve T cells. Responses included several CD4⁺ and CD8⁺ T cell epitopes shown to be recognized in vivo by acutely and chronically infected individuals and some CD4⁺ T cell epitopes not identified previously. Follow up studies of acute and recent HIV infected samples revealed that these latter epitopes are among the earliest recognized in vivo, but the responses are lost rapidly, presumably through activation-induced general CD4⁺ T cell depletion which renders the newly activated HIV-specific CD4⁺ T cells prime targets for elimination.

Conclusion

Our studies highlight the ability of DCs to efficiently prime naïve T cells and induce a broad repertoire of HIV-specific responses and also provide valuable insights to the pathogenesis of HIV-1 infection in vivo.