A key event in the generation of a cellular response against malicious organisms through the endocytic pathway is binding of peptidic antigens by major histocompatibility complex class II (MHC class II) molecules. The bound peptide is then presented on the cell surface where it can be recognized by T helper lymphocytes. NetMHCIIpan is a state-of-the-art method for the quantitative prediction of peptide binding to any human or mouse MHC class II molecule of known sequence. In this paper, we describe an updated version of the method with improved peptide binding register identification. Binding register prediction is concerned with determining the minimal core region of nine residues directly in contact with the MHC binding cleft, a crucial piece of information both for the identification and design of CD4+ T cell antigens. When applied to a set of 51 crystal structures of peptide-MHC complexes with known binding registers, the new method NetMHCIIpan-3.1 significantly outperformed the earlier 3.0 version. We illustrate the impact of accurate binding core identification for the interpretation of T cell cross-reactivity using tetramer double staining with a CMV epitope and its variants mapped to the epitope binding core. NetMHCIIpan is publicly available at http://www.cbs.dtu.dk/services/NetMHCIIpan-3.1.
MHC class II; Peptide binding; Tcell cross-reactivity; Binding core; Artificial neural networks; Peptide-MHC
CD8+ T cells contribute to the control of HIV, but it is not clear whether initial immune responses modulate the viral set point. We screened high-risk uninfected women twice a week for plasma HIV RNA and identified twelve hyperacute infections. Onset of viremia elicited a massive HIV-specific CD8+ T cell response, with limited bystander activation of non-HIV memory CD8+ T cells. HIV-specific CD8+ T cells secreted little interferon-γ, underwent rapid apoptosis and failed to upregulate the interleukin 7 receptor, known to be important for T cell survival. The rapidity to peak CD8+ T cell activation and the absolute magnitude of activation induced by the exponential rise in viremia were inversely correlated with set point viremia. These data indicate that rapid, high magnitude HIV-induced CD8+ T cell responses are crucial for subsequent immune control of acute infection, which has important implications for HIV vaccine design.
Tumour cells can evade the immune system by dysregulation of human leukocyte antigens (HLA-I). Low quantity and/or altered quality of HLA-I cell surface expression is the result of either HLA-I alterations or dysregulations of proteins of the antigen-processing machinery (APM). Tapasin is an APM protein dedicated to the maturation of HLA-I and dysregulation of tapasin has been linked to higher malignancy in several different tumours.
We studied the expression of APM components and HLA-I, as well as HLA-I tapasin-dependency profiles in glioblastoma tissues and corresponding cell lines.
Tapasin displayed the strongest correlation to HLA-I heavy chain but also clustered with β2-microglobulin, transporter associated with antigen processing (TAP) and LMP. Moreover, tapasin also correlated to survival of glioblastoma patients. Some APM components, for example, TAP1/TAP2 and LMP2/LMP7, showed variable but coordinated expression, whereas ERAP1/ERAP2 displayed an imbalanced expression pattern. Furthermore, analysis of HLA-I profiles revealed variable tapasin dependence of HLA-I allomorphs in glioblastoma patients.
Expression of APM proteins is highly variable between glioblastomas. Tapasin stands out as the APM component strongest correlated to HLA-I expression and we proved that HLA-I profiles in glioblastoma patients include tapasin-dependent allomorphs. The level of tapasin was also correlated with patient survival time. Our results support the need for individualisation of immunotherapy protocols.
tapasin; HLA-I; glioblastoma multiforme; peptide; tapasin-dependency; immunotherapy
The live attenuated yellow fever vaccine (YF-17D) has been successfully used for more than 70 years. It is generally considered a safe vaccine, however, recent reports of serious adverse events following vaccination have raised concerns and led to suggestions that even safer YF vaccines should be developed. Replication deficient adenoviruses (Ad) have been widely evaluated as recombinant vectors, particularly in the context of prophylactic vaccination against viral infections in which induction of CD8+ T-cell mediated immunity is crucial, but potent antibody responses may also be elicited using these vectors. In this study, we present two adenobased vectors targeting non-structural and structural YF antigens and characterize their immunological properties. We report that a single immunization with an Ad-vector encoding the non-structural protein 3 from YF-17D could elicit a strong CD8+ T-cell response, which afforded a high degree of protection from subsequent intracranial challenge of vaccinated mice. However, full protection was only observed using a vector encoding the structural proteins from YF-17D. This vector elicited virus-specific CD8+ T cells as well as neutralizing antibodies, and both components were shown to be important for protection thus mimicking the situation recently uncovered in YF-17D vaccinated mice. Considering that Ad-vectors are very safe, easy to produce and highly immunogenic in humans, our data indicate that a replication deficient adenovector-based YF vaccine may represent a safe and efficient alternative to the classical live attenuated YF vaccine and should be further tested.
Live attenuated yellow fever vaccine (YF-17D) is an efficient and generally safe vaccine. Nevertheless, in recent years the reporting of serious adverse effects together with the given limitations in the use of this live vaccine in certain risk groups has spurred an interest in developing a more generally applicable and safer alternative. Using an adenovector platform and recombinant vaccines targeting both structural and non-structural YF antigens, we now demonstrate that non-replicating adenobased vaccines may be used to induce a state of host immunity, which like YF-17D vaccination encompasses both major arms of the adaptive immune system. Furthermore, in a murine challenge model, adenovector induced protection fully matched that induced by the current vaccine. Taken together our results strongly suggest that adenovectored vaccines targeting structural and non-structural viral antigens represent a viable and safe alternative to the existing live, attenuated YF vaccine.
The attenuated yellow fever vaccine (YF-17D) was developed in the 1930’s, yet little is known about the protective mechanisms underlying its efficiency. In this study, we analyzed the relative contribution of cell-mediated and humoral immunity to the vaccine-induced protection in a murine model of YF-17D infection. Using different strains of knock-out mice, we found that CD4+ T cells, B cells and antibodies are required for full clinical protection of vaccinated mice, while CD8+ T cells are dispensable for long-term survival following intracerebral (i.c.) challenge. However, by analysing the immune response inside the infected CNS, we observed an accelerated T-cell influx into the brain following i.c. challenge of vaccinated mice, and this T-cell recruitment correlated with improved virus control in the brain. Using mice deficient in B cells we found that, in the absence of antibodies, YF vaccination can still induce some antiviral protection, and in vivo depletion of CD8+ T cells from these animals revealed a pivotal role for CD8+ T cells in controlling virus replication in the absence of a humoral response. Finally, we demonstrated that effector CD8+ T cells also contribute to viral control in the presence of circulating YF-specific antibodies. To our knowledge this is the first time that YF-specific CD8+ T cells have been demonstrated to possess antiviral activity in vivo.
The aim of the present study was to identify influenza A-derived peptides which bind to both HLA class I and -II molecules and by immunization lead to both HLA class I and class II restricted immune responses. Eight influenza A-derived 9-11mer peptides with simultaneous binding to both HLA-A*02:01 and HLA-DRB1*01:01 molecules were identified by bioinformatics and biochemical technology. Immunization of transgenic HLA-A*02:01/HLA-DRB1*01:01 mice with four of these double binding peptides gave rise to both HLA class I and class II restricted responses by CD8 and CD4 T cells, respectively, whereas four of the double binding peptides did result in HLA-A*02:01 restricted responses only. According to their cytokine profile, the CD4 T cell responses were of the Th2 type. In influenza infected mice, we were unable to detect natural processing in vivo of the double restricted peptides and in line with this, peptide vaccination did not decrease virus titres in the lungs of intranasally influenza challenged mice. Our data show that HLA class I and class II double binding peptides can be identified by bioinformatics and biochemical technology. By immunization, double binding peptides can give rise to both HLA class I and class I restricted responses, a quality which might be of potential interest for peptide-based vaccine development.
Immunodominance describes a phenomenon whereby the immune system consistently targets only a fraction of the available antigen pool derived from a given pathogen. In the case of CD8+ T-cells, these constrained epitope targeting patterns are linked to human leukocyte antigen (HLA) class-I expression and determine disease progression. Despite the biological importance of these predetermined response hierarchies, however, little is known about the factors that control immunodominance in vivo. In this study, we conducted an extensive analysis of CD8+ T-cell responses restricted by a single HLA class-I molecule to evaluate the mechanisms that contribute to epitope targeting frequency and antiviral efficacy in HIV-1 infection. A clear immunodominance hierarchy was observed across 20 different epitopes restricted by HLA-B*42:01, which is highly prevalent in populations of African origin. Moreover, in line with previous studies, Gag-specific responses and targeting breadth were associated with lower viral load set-points. However, peptide-HLA-B*42:01 binding affinity and stability were not significantly linked with targeting frequencies. Instead, immunodominance correlated with epitope-specific usage of public TCRs, defined as amino acid residue-identical TRB sequences that occur in multiple individuals. Collectively, these results provide the first insights into a potential link between shared TCR recruitment, immunodominance and antiviral efficacy in a major human infection.
CD8+ T-cell; clonotype; HIV-1; immunodominance; TCR
The binding of peptides to classical major histocompatibility complex (MHC) class I proteins is the single most selective step in antigen presentation. However, the peptide-binding specificity of cattle MHC (bovine leucocyte antigen, BoLA) class I (BoLA-I) molecules remains poorly characterized. Here, we demonstrate how a combination of high-throughput assays using positional scanning combinatorial peptide libraries, peptide dissociation, and peptide-binding affinity binding measurements can be combined with bioinformatics to effectively characterize the functionality of BoLA-I molecules. Using this strategy, we characterized eight BoLA-I molecules, and found the peptide specificity to resemble that of human MHC-I molecules with primary anchors most often at P2 and P9, and occasional auxiliary P1/P3/P5/P6 anchors. We analyzed nine reported CTL epitopes from Theileria parva, and in eight cases, stable and high affinity binding was confirmed. A set of peptides were tested for binding affinity to the eight BoLA proteins and used to refine the predictors of peptide–MHC binding NetMHC and NetMHCpan. The inclusion of BoLA-specific peptide-binding data led to a significant improvement in prediction accuracy for reported T. parva CTL epitopes. For reported CTL epitopes with weak or no predicted binding, these refined prediction methods suggested presence of nested minimal epitopes with high-predicted binding affinity. The enhanced affinity of the alternative peptides was in all cases confirmed experimentally. This study demonstrates how biochemical high-throughput assays combined with immunoinformatics can be used to characterize the peptide-binding motifs of BoLA-I molecules, boosting performance of MHC peptide-binding prediction methods, and empowering rational epitope discovery in cattle.
Bovine leucocyte antigen; BoLA; Rational epitope discovery; CTL epitopes; Immunoinformatics
Immunotherapy has increased overall survival of metastatic cancer patients, and cancer antigens are promising vaccine targets. To fulfill the promise, appropriate tailoring of the vaccine formulations to mount in vivo cytotoxic T cell (CTL) responses toward co-delivered cancer antigens is essential. Previous development of therapeutic cancer vaccines has largely been based on studies in mice, and the majority of these candidate vaccines failed to induce therapeutic responses in the subsequent human clinical trials. Given that antigen dose and vaccine volume in pigs are translatable to humans and the porcine immunome is closer related to the human counterpart, we here introduce pigs as a supplementary large animal model for human cancer vaccine development. IDO and RhoC, both important in human cancer development and progression, were used as vaccine targets and 12 pigs were immunized with overlapping 20mer peptides spanning the entire porcine IDO and RhoC sequences formulated in CTL-inducing adjuvants: CAF09, CASAC, Montanide ISA 51 VG, or PBS. Taking advantage of recombinant swine MHC class I molecules (SLAs), the peptide-SLA complex stability was measured for 198 IDO- or RhoC-derived 9-11mer peptides predicted to bind to SLA-1*04:01, −1*07:02, −2*04:01, −2*05:02, and/or −3*04:01. This identified 89 stable (t½ ≥ 0.5 h) peptide-SLA complexes. By IFN-γ release in PBMC cultures we monitored the vaccine-induced peptide-specific CTL responses, and found responses to both IDO- and RhoC-derived peptides across all groups with no adjuvant being superior. These findings support the further use of pigs as a large animal model for vaccine development against human cancer.
immune therapy; cancer vaccines; cytotoxic T cells; animal model; peptide-MHC stability; adjuvants; immunologic
Genetic polymorphisms within the MHC encoding region have the strongest impact on HIV disease progression of any in the human genome and provide important clues to the mechanisms of HIV immune control. Few analyses have been undertaken of HLA alleles associated with rapid disease progression. HLA-B*07:02 is an HLA class I molecule that is prevalent in most populations worldwide and that has previously been consistently linked to accelerated disease progression in B-clade infection. This study investigates the observation that HLA-B*07:02 is not associated with a high viral setpoint in C-clade infection. We examine the hypothesis that this clade-specific difference in association with disease outcome may be related to distinct targeting of CD8+ T cell epitopes. We observed that C-clade-infected individuals with HLA-B*07:02 target a broader range of Gag epitopes, and to higher magnitudes, than do individuals infected with B-clade infection. In particular, a novel p17-Gag (Gag22-30, RPGGKKHYM) epitope is targeted in >50% of HLA-B*07:02-positive C-clade-infected individuals but clade-specific differences in this epitope result in nonimmunogenicity in B-clade infection. Only the C-clade p24-Gag “GL9” (Gag355-363, GPSHKARVL) epitope-specific CD8+ T cell response out of 16 studied was associated with a low viral setpoint. Although this epitope was also targeted in B-clade infection, the escape mutant S357S is present at higher frequency in B-clade infection than in C-clade infection (70% versus 43% in HLA-B*07:02-negative subjects). These data support earlier studies suggesting that increased breadth of the Gag-specific CD8+ T cell response may contribute to improved HIV immune control irrespective of the particular HLA molecules expressed.
Highly polymorphic major histocompatibility complex (MHC) molecules are at the heart of adaptive immune responses, playing crucial roles in many kinds of disease and in vaccination. We report that breadth of peptide presentation and level of cell surface expression of class I molecules are inversely correlated in both chickens and humans. This relationship correlates with protective responses against infectious pathogens including Marek's disease virus leading to lethal tumours in chickens and human immunodeficiency virus infection progressing to AIDS in humans. We propose that differences in peptide binding repertoire define two groups of MHC class I molecules strategically evolved as generalists and specialists for different modes of pathogen resistance. We suggest that differences in cell surface expression level ensure the development of optimal peripheral T cell responses. The inverse relationship of peptide repertoire and expression is evidently a fundamental property of MHC molecules, with ramifications extending beyond immunology and medicine to evolutionary biology and conservation.
Our immune system has the remarkable ability to recognize and destroy damaged cells or invading microbes while leaving the healthy cells in the body alone. Groups of proteins called ‘MHC class I molecules’ play important roles in defence against invading microbes. If a cell becomes infected with a virus or a bacterium, these molecules can recognize and bind to fragments of foreign or unusual proteins inside the cell, and display them on the surface of the cell. This allows immune cells to identify and kill the infected cells.
Cells produce many different MHC class I molecules that are able to bind to different protein fragments. Some MHC molecules can bind to a wider variety of protein fragments than others, and the number of these molecules present on the cell surface can also vary between individuals. Researchers have noticed that individuals with particular MHC molecules tend to be more or less resistant to particular diseases. For instance, individuals with certain MHC molecules tend to take longer to develop AIDS if they become infected with HIV. However, it is not clear how either the variety of protein fragments bound or the numbers of MHC class I molecules on the surface of cells could alter the immune response.
Here, Chappell, Meziane et al. studied MHC class I molecules in chickens and humans. The experiments reveal that the MHC class I molecules that can bind to a larger variety of protein fragments (so-called ‘generalists’) are present in lower numbers on the surface of cells than molecules that can bind to a smaller variety of fragments (so-called ‘specialists’).
Furthermore, generalist MHC molecules were found to provide resistance to Marek's disease in chickens—which causes paralysis—but some specialists slowed the progression of HIV infections into AIDS in humans. Chappell, Meziane et al. propose that these two types of MHC class I molecules evolved to perform different roles in immune responses. This is a new way of looking at the role of MHC molecules in fighting disease, and the next challenge is to explore the implications for medicine and evolutionary biology.
HIV; Marek's disease; peptide repertoire; avian; MHC; chicken; human
Efficient presentation of peptide-MHC class I complexes to immune T cells depends upon stable peptide-MHC class I interactions. Theoretically, determining the rate of dissociation of a peptide-MHC class I complexes is straightforward; in practical terms, however, generating the accurate and closely timed data needed to determine the rate of dissociation is not simple. Ideally, one should use a homogenous assay involving an inexhaustible and label-free assay principle. Here, we present a homogenous, high-throughput peptide-MHC class I dissociation assay, which by and large fulfill these ideal requirements. To avoid labeling of the highly variable peptide, we labeled the invariant β2m and monitored its dissociation by a scintillation proximity assay, which has no separation steps and allows for real-time quantitative measurement of dissociation. Validating this work-around to create a virtually label-free assay, we showed that rates of peptide-MHC class I dissociation measured in this assay correlated well with rates of dissociation rates measured conventionally with labeled peptides. This assay can be used to measure the stability of any peptide-MHC Class I combination, it is reproducible and it is well suited for high-throughput screening. To exemplify this, we screened a panel of 384 high-affinity peptides binding to the MHC class I molecule, HLA-A*02:01, and observed rates of dissociation that ranged from 0.1 hours to 46 hours depending on the peptide used.
MHC-I; peptide; stability; dissociation; scintillation proximity assay
Presentation of identical HIV-1 peptides by closely related Human Leukocyte Antigen class I (HLAI) molecules can select distinct patterns of escape mutation that have a significant impact on viral fitness and disease progression. The molecular mechanisms by which HLAI micropolymorphisms can induce differential HIV-1 escape patterns within identical peptide epitopes remain unknown.
Here, we undertook genetic and structural analyses of two immunodominant HIV-1 peptides, Gag180–188 (TPQDLNTML, TL9-p24) and Nef71–79 (RPQVPLRPM, RM9-Nef) that are among the most highly targeted epitopes in the global HIV-1 epidemic. We show that single polymorphisms between different alleles of the HLA-B7 superfamily can induce a conformational switch in peptide conformation that is associated with differential HLAI-specific escape mutation and immune control. A dominant R71K mutation in the Nef71-79 occurred in those with HLA-B*07:02 but not B*42:01/02 or B*81:01. No structural difference in the HLA-epitope complexes was detected to explain this observation.
These data suggest that identical peptides presented through very similar HLAI landscapes are recognized as distinct epitopes and provide a novel structural mechanism for previously observed differential HIV-1 escape and disease progression.
Electronic supplementary material
The online version of this article (doi:10.1186/s12977-015-0149-5) contains supplementary material, which is available to authorized users.
HIV-1; CD8+ T-Cells; Viral escape; pHLA structure
Fifty-five percent of individuals with HLA-B*57:01 exposed to the antiretroviral drug abacavir develop a hypersensitivity reaction (HSR) that has been attributed to naïve T-cell responses to neo-antigen generated by the drug. Immunologically confirmed abacavir HSR can manifest clinically in less than 48 hours following first exposure suggesting that, at least in some cases, abacavir HSR is due to re-stimulation of a pre-existing memory T-cell population rather than priming of a high frequency naïve T-cell population.
To determine whether a pre-existing abacavir reactive memory T-cell population contributes to early abacavir HSR symptoms, we studied the abacavir specific naïve or memory T-cell response using HLA-B*57:01 positive HSR patients or healthy controls using ELISpot assay, intra-cellular cytokine staining and tetramer labelling.
Abacavir reactive CD8+ T-cell responses were detected in vitro in one hundred percent of abacavir unexposed HLA-B*57:01 positive healthy donors. Abacavir-specific CD8+ T cells from such donors can be expanded from sorted memory, and sorted naïve, CD8+ T cells without need for autologous CD4+ T cells.
We propose that these pre-existing abacavir-reactive memory CD8+ T-cell responses must have been primed by earlier exposure to another foreign antigen and that these T cells cross-react with an abacavir-HLA-B*57:01-endogenous peptide ligand complex, in keeping with the model of heterologous immunity proposed in transplant rejection.
Celiac disease is caused by intolerance to cereal gluten proteins, and HLA-DQ molecules are involved in the disease pathogenesis by presentation of gluten peptides to CD4+ T cells. The α- or β-chain sharing HLA molecules DQ2.5, DQ2.2, and DQ7.5 display different risks for the disease. It was recently demonstrated that T cells of DQ2.5 and DQ2.2 patients recognize distinct sets of gluten epitopes, suggesting that these two DQ2 variants select different peptides for display. To explore whether this is the case, we performed a comprehensive comparison of the endogenous self-peptides bound to HLA-DQ molecules of B-lymphoblastoid cell lines. Peptides were eluted from affinity-purified HLA molecules of nine cell lines and subjected to quadrupole orbitrap mass spectrometry and MaxQuant software analysis. Altogether, 12,712 endogenous peptides were identified at very different relative abundances. Hierarchical clustering of normalized quantitative data demonstrated significant differences in repertoires of peptides between the three DQ variant molecules. The neural network-based method, NNAlign, was used to identify peptide-binding motifs. The binding motifs of DQ2.5 and DQ7.5 concurred with previously established binding motifs. The binding motif of DQ2.2 was strikingly different from that of DQ2.5 with position P3 being a major anchor having a preference for threonine and serine. This is notable as three recently identified epitopes of gluten recognized by T cells of DQ2.2 celiac patients harbor serine at position P3. This study demonstrates that relative quantitative comparison of endogenous peptides sampled from our protein metabolism by HLA molecules provides clues to understand HLA association with disease.
Electronic supplementary material
The online version of this article (doi:10.1007/s00251-014-0819-9) contains supplementary material, which is available to authorized users.
Antigen presentation/processing; Binding motif; Celiac disease; Mass spectrometry; MHC
MHC class I molecules (HLA-I in humans) present peptides derived from endogenous proteins to CTLs. Whereas the peptide-binding specificities of HLA-A and -B molecules have been studied extensively, little is known about HLA-C specificities. Combining a positional scanning combinatorial peptide library approach with a peptide–HLA-I dissociation assay, in this study we present a general strategy to determine the peptide-binding specificity of any MHC class I molecule. We applied this novel strategy to 17 of the most common HLA-C molecules, and for 16 of these we successfully generated matrices representing their peptide-binding motifs. The motifs prominently shared a conserved C-terminal primary anchor with hydrophobic amino acid residues, as well as one or more diverse primary and auxiliary anchors at P1, P2, P3, and/or P7. Matrices were used to generate a large panel of HLA-C–specific peptide-binding data and update our pan-specific NetMHCpan predictor, whose predictive performance was considerably improved with respect to peptide binding to HLA-C. The updated predictor was used to assess the specificities of HLA-C molecules, which were found to cover a more limited sequence space than HLA-A and -B molecules. Assessing the functional significance of these new tools, HLA-C*07:01 transgenic mice were immunized with stable HLA-C*07:01 binders; six of six tested stable peptide binders were immunogenic. Finally, we generated HLA-C tetramers and labeled human CD8+ T cells and NK cells. These new resources should support future research on the biology of HLA-C molecules. The data are deposited at the Immune Epitope Database, and the updated NetMHCpan predictor is available at the Center for Biological Sequence Analysis and the Immune Epitope Database.
Major histocompatibility complex class II (MHCII) molecules play an important role in cell-mediated immunity. They present specific peptides derived from endosomal proteins for recognition by T helper cells. The identification of peptides that bind to MHCII molecules is therefore of great importance for understanding the nature of immune responses and identifying T cell epitopes for the design of new vaccines and immunotherapies. Given the large number of MHC variants, and the costly experimental procedures needed to evaluate individual peptide–MHC interactions, computational predictions have become particularly attractive as first-line methods in epitope discovery. However, only a few so-called pan-specific prediction methods capable of predicting binding to any MHC molecule with known protein sequence are currently available, and all of them are limited to HLA-DR. Here, we present the first pan-specific method capable of predicting peptide binding to any HLA class II molecule with a defined protein sequence. The method employs a strategy common for HLA-DR, HLA-DP and HLA-DQ molecules to define the peptide-binding MHC environment in terms of a pseudo sequence. This strategy allows the inclusion of new molecules even from other species. The method was evaluated in several benchmarks and demonstrates a significant improvement over molecule-specific methods as well as the ability to predict peptide binding of previously uncharacterised MHCII molecules. To the best of our knowledge, the NetMHCIIpan-3.0 method is the first pan-specific predictor covering all HLA class II molecules with known sequences including HLA-DR, HLA-DP, and HLA-DQ. The NetMHCpan-3.0 method is available at http://www.cbs.dtu.dk/services/NetMHCIIpan-3.0.
MHC class II; Tcell epitope; MHC binding specificity; Peptide–MHC binding; Human leukocyte antigens; Artificial neural networks
The identification of peptides binding to major histocompatibility complexes (MHC) is a critical step in the understanding of T cell immune responses. The human MHC genomic region (HLA) is extremely polymorphic comprising several thousand alleles, many encoding a distinct molecule. The potentially unique specificities remain experimentally uncharacterized for the vast majority of HLA molecules. Likewise, for nonhuman species, only a minor fraction of the known MHC molecules have been characterized. Here, we describe a tool, MHCcluster, to functionally cluster MHC molecules based on their predicted binding specificity. The method has a flexible web interface that allows the user to include any MHC of interest in the analysis. The output consists of a static heat map and graphical tree-based visualizations of the functional relationship between MHC variants and a dynamic TreeViewer interface where both the functional relationship and the individual binding specificities of MHC molecules are visualized. We demonstrate that conventional sequence-based clustering will fail to identify the functional relationship between molecules, when applied to MHC system, and only through the use of the predicted binding specificity can a correct clustering be found. Clustering of prevalent HLA-A and HLA-B alleles using MHCcluster confirms the presence of 12 major specificity groups (supertypes) some however with highly divergent specificities. Importantly, some HLA molecules are shown not to fit any supertype classification. Also, we use MHCcluster to show that chimpanzee MHC class I molecules have a reduced functional diversity compared to that of HLA class I molecules. MHCcluster is available at www.cbs.dtu.dk/services/MHCcluster-2.0.
MHC; HLA; Binding motif; Functional clustering; MHC specificity; Supertypes
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.
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.
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.
Benchmarking of MHC class I predictors; Epitope prediction; Sequence similarity; Cross-validation
We have generated a panel of transgenic mice expressing HLA-A*01:03, -A*24:02, -B*08:01, -B*27:05, -B*35:01, -B*44:02, or -C*07:01 as chimeric monochain molecules (i.e., appropriate HLA α1α2 H chain domains fused with a mouse α3 domain and covalently linked to human β2-microglobulin). Whereas surface expression of several transgenes was markedly reduced in recipient mice that coexpressed endogenous H-2 class I molecules, substantial surface expression of all human transgenes was observed in mice lacking H-2 class I molecules. In these HLA monochain transgenic/H-2 class I null mice, we observed a quantitative and qualitative restoration of the peripheral CD8+ T cell repertoire, which exhibited a TCR diversity comparable with C57BL/6 WT mice. Potent epitope-specific, HLA-restricted, IFN-γ–producing CD8+ T cell responses were generated against known reference T cell epitopes after either peptide or DNA immunization. HLA-wise, these new transgenic strains encompass a large proportion of individuals from all major human races and ethnicities. In combination with the previously created HLA-A*02:01 and -B*07:02 transgenic mice, the novel HLA transgenic mice described in this report should be a versatile preclinical animal model that will speed up the identification and optimization of HLA-restricted CD8+ T cell epitopes of potential interest in various autoimmune human diseases and in preclinical evaluation of T cell–based vaccines.
Human cytomegalovirus (HCMV) is an important human pathogen. It is a leading cause of congenital infection and a leading infectious threat to recipients of solid organ transplants as well as of allogeneic hematopoietic cell transplants. Moreover, it has recently been suggested that HCMV may promote tumor development. Both CD4+ and CD8+ T cell responses are important for long-term control of the virus, and adoptive transfer of HCMV-specific T cells has led to protection from reactivation and HCMV disease. Identification of HCMV-specific T cell epitopes has primarily focused on CD8+ T cell responses against the pp65 phosphoprotein. In this study, we have focused on CD4+ and CD8+ T cell responses against the immediate early 1 and 2 proteins (IE1 and IE2). Using overlapping peptides spanning the entire IE1 and IE2 sequences, peripheral blood mononuclear cells from 16 healthy, HLA-typed, donors were screened by ex vivo IFN-γ ELISpot and in vitro intracellular cytokine secretion assays. The specificities of CD4+ and CD8+ T cell responses were identified and validated by HLA class II and I tetramers, respectively. Eighty-one CD4+ and 44 CD8+ T cell responses were identified representing at least seven different CD4 epitopes and 14 CD8 epitopes restricted by seven and 11 different HLA class II and I molecules, respectively, in total covering 91 and 98% of the Caucasian population, respectively. Presented in the context of several different HLA class II molecules, two epitope areas in IE1 and IE2 were recognized in about half of the analyzed donors. These data may be used to design a versatile anti-HCMV vaccine and/or immunotherapy strategy.
HLA-B*57 is strongly associated with immune control of HIV and delayed AIDS progression. The closely related, but less protective, HLA-B*58:01 presents similar epitopes, but HLA-B*58:01+ individuals do not generate CD8+ T cells targeting the KF11-Gag epitope, which has been linked to low viremia. Here we show that HLA-B*58:01 binds and presents KF11 peptide, but HIV-infected HLA-B*58:01+ cells fail to process KF11. This unexpected finding demonstrates that immunodominance patterns can be influenced by intracellular events independent of HLA binding motifs.
Peptide-major histocompatibility complex (p-MHC) class I tetramer complexes have facilitated the early detection and functional characterisation of epitope specific CD8+ cytotoxic T lymphocytes (CTL). Here, we report on the generation of seven recombinant bovine leukocyte antigens (BoLA) and recombinant bovine β2-microglobulin from which p-MHC class I tetramers can be derived in ~48 h. We validated a set of p-MHC class I tetramers against a panel of CTL lines specific to seven epitopes on five different antigens of Theileria parva, a protozoan pathogen causing the lethal bovine disease East Coast fever. One of the p-MHC class I tetramers was tested in ex vivo assays and we detected T. parva specific CTL in peripheral blood of cattle at day 15-17 post-immunization with a live parasite vaccine. The algorithm NetMHCpan predicted alternative epitope sequences for some of the T. parva CTL epitopes. Using an ELISA assay to measure peptide-BoLA monomer formation and p-MHC class I tetramers of new specificity, we demonstrate that a predicted alternative epitope Tp229-37 rather than the previously reported Tp227-37 epitope is the correct Tp2 epitope presented by BoLA-6*04101. We also verified the prediction by NetMHCpan that the Tp587-95 epitope reported as BoLA-T5 restricted can also be presented by BoLA-1*02301, a molecule similar in sequence to BoLA-T5. In addition, Tp587-95 specific bovine CTL were simultaneously stained by Tp5-BoLA-1*02301 and Tp5-BoLA-T5 tetramers suggesting that one T cell receptor can bind to two different BoLA MHC class I molecules presenting the Tp587-95 epitope and that these BoLA molecules fall into a single functional supertype.