A transfer-learning approach to predict antigen immunogenicity and T-cell receptor specificity

Antigen immunogenicity and the specificity of binding of T-cell receptors to antigens are key properties underlying effective immune responses. Here we propose diffRBM, an approach based on transfer learning and Restricted Boltzmann Machines, to build sequence-based predictive models of these properties. DiffRBM is designed to learn the distinctive patterns in amino-acid composition that, on the one hand, underlie the antigen's probability of triggering a response, and on the other hand the T-cell receptor's ability to bind to a given antigen. We show that the patterns learnt by diffRBM allow us to predict putative contact sites of the antigen-receptor complex. We also discriminate immunogenic and non-immunogenic antigens, antigen-specific and generic receptors, reaching performances that compare favorably to existing sequence-based predictors of antigen immunogenicity and T-cell receptor specificity.

No Difference in the Prevalence of HIV-1 gag Cytotoxic T-Lymphocyte-Associated Escape Mutations in Viral Sequences from Early and Late Parts of the HIV-1 Subtype C Pandemic in Botswana

HIV is known to accumulate escape mutations in the gag gene in response to the immune response from cytotoxic T lymphocytes (CTLs). These mutations can occur within an individual as well as at a population level. The population of Botswana exhibits a high prevalence of HLA*B57 and HLA*B58, which are associated with effective immune control of HIV. In this retrospective cross-sectional investigation, HIV-1 gag gene sequences were analyzed from recently infected participants across two time periods which were 10 years apart: the early time point (ETP) and late time point (LTP). The prevalence of CTL escape mutations was relatively similar between the two time points-ETP (10.6%) and LTP (9.7%). The P17 protein had the most mutations (9.4%) out of the 36 mutations that were identified. Three mutations (A83T, K18R, Y79H) in P17 and T190A in P24 were unique to the ETP sequences at a prevalence of 2.4%, 4.9%, 7.3%, and 5%, respectively. Mutations unique to the LTP sequences were all in the P24 protein, including T190V (3%), E177D (6%), R264K (3%), G248D (1%), and M228L (11%). Mutation K331R was statistically higher in the ETP (10%) compared to the LTP (1%) sequences (p < 0.01), while H219Q was higher in the LTP (21%) compared to the ETP (5%) (p < 0.01). Phylogenetically, the gag sequences clustered dependently on the time points. We observed a slower adaptation of HIV-1C to CTL immune pressure at a population level in Botswana. These insights into the genetic diversity and sequence clustering of HIV-1C can aid in the design of future vaccine strategies.

Altered MUC1 epitope-specific CTLs: A potential target for immunotherapy of pancreatic cancer

The efficacy of conventional treatments for pancreatic cancer remains unsatisfactory, and immunotherapy is an emerging option for adjuvant treatment of this highly deadly disorder. The tumor-associated antigen (TAA) MUC1 is expressed in a variety of human cancers and is overexpressed in more than 90% of pancreatic cancer, which makes it an attractive target for cancer immunotherapy. As a self-protein, MUC1 shows a low immunogenicity because of immune tolerance, and the most effective approach to breaking immune tolerance is alteration of the antigen structure. In this study, the altered MUC1_1068-1076Y1 epitope (YLQRDISEM) by modification of amino acid residues in sequences presented a higher immunogenicity and elicited more CTLs relative to the wild-type (WT) MUC1_1068-1076 epitope (ELQRDISEM). In addition, the altered MUC1_1068-1076Y1 epitope was found to cross-recognize pancreatic cancer cells expressing WT MUC1 peptides in an HLA-A0201-restricted manner and trigger stronger immune responses against pancreatic cancer via the perforin/granzyme apoptosis pathway. As a potential HLA-A0201-restricted CTL epitope, the altered MUC1_1068-1076Y1 epitope is considered as a promising target for immunotherapy of pancreatic cancer. Alteration of epitope residues may be feasible to solve the problem of the low immunogenicity of TAA and break immune tolerance to induce immune responses against human cancers.

In vivo therapeutic effects of affinity-improved-TCR engineered T-cells on HBV-related hepatocellular carcinoma

Background

In patients with hepatitis B virus (HBV)-related hepatocellular carcinoma (HCC), virus-specific cytotoxic T lymphocytes (CTLs) fail to eliminate HCC cells expressing HBV antigens. As the expression of viral antigen in HBV-associated HCC may decrease to allow tumor to escape immune attacks, we hypothesized that an HBV surface antigen (HBsAg)-specific affinity-improved-T-cell receptor (TCR) will enable T cells to target HCC more effectively than corresponding wild-type-TCR. We also postulated that TCR promiscuity can be exploited to efficiently capture HBV variants that can hinder CTL-based therapeutics.

Methods

We applied flexi-panning to isolate affinity-improved TCRs binding to a variant antigen, the human leukocyte antigen (HLA)-A*02:01-restricted nonapeptide HBs_371-379-ILSPFLPLL, from libraries constructed with a TCR cloned using the decapeptide HBs_370-379-SIVSPFIPLL. The potency and safety of the affinity-improved-TCR engineered T-cells (Ai-TCR-T) were verified with potentially cross-reactive human and HBV-variant peptides, tumor and normal cells, and xenograft mouse models.

Results

Ai-TCR-T cells retained cognate HBV antigen specificity and recognized a wide range of HBV genotypic variants with improved sensitivity and cytotoxicity. Cell infusions produced complete elimination of HCC without recurrence in the xenograft mouse models. Elevated accumulation of CD8⁺ Ai-TCR-T cells in tumors correlated with tumor shrinkage.

Conclusion

The in vitro and in vivo studies demonstrated that HBsAg-specific Ai-TCR-T cells had safety profiles similar to those of their wild-type counterparts and significantly enhanced potency. This study presents an approach to develop new therapeutic strategies for HBV-related HCC.

A de novo approach to inferring within-host fitness effects during untreated HIV-1 infection

In the absence of effective antiviral therapy, HIV-1 evolves in response to the within-host environment, of which the immune system is an important aspect. During the earliest stages of infection, this process of evolution is very rapid, driven by a small number of CTL escape mutations. As the infection progresses, immune escape variants evolve under reduced magnitudes of selection, while competition between an increasing number of polymorphic alleles (i.e., clonal interference) makes it difficult to quantify the magnitude of selection acting upon specific variant alleles. To tackle this complex problem, we developed a novel multi-locus inference method to evaluate the role of selection during the chronic stage of within-host infection. We applied this method to targeted sequence data from the p24 and gp41 regions of HIV-1 collected from 34 patients with long-term untreated HIV-1 infection. We identify a broad distribution of beneficial fitness effects during infection, with a small number of variants evolving under strong selection and very many variants evolving under weaker selection. The uniquely large number of infections analysed granted a previously unparalleled statistical power to identify loci at which selection could be inferred to act with statistical confidence. Our model makes no prior assumptions about the nature of alleles under selection, such that any synonymous or non-synonymous variant may be inferred to evolve under selection. However, the majority of variants inferred with confidence to be under selection were non-synonymous in nature, and in most cases were have previously been associated with either CTL escape in p24 or neutralising antibody escape in gp41. We also identified a putative new CTL escape site (residue 286 in gag), and a region of gp41 (including residues 644, 648, 655 in env) likely to be associated with immune escape. Sites inferred to be under selection in multiple hosts have high within-host and between-host diversity although not all sites with high between-host diversity were inferred to be under selection at the within-host level. Our identification of selection at sites associated with resistance to broadly neutralising antibodies (bNAbs) highlights the need to fully understand the role of selection in untreated individuals when designing bNAb based therapies.

Identifying the immune interactions underlying HLA class I disease associations

Variation in the risk and severity of many autoimmune diseases, malignancies and infections is strongly associated with polymorphisms at the HLA class I loci. These genetic associations provide a powerful opportunity for understanding the etiology of human disease. HLA class I associations are often interpreted in the light of 'protective' or 'detrimental' CD8⁺ T cell responses which are restricted by the host HLA class I allotype. However, given the diverse receptors which are bound by HLA class I molecules, alternative interpretations are possible. As well as binding T cell receptors on CD8⁺ T cells, HLA class I molecules are important ligands for inhibitory and activating killer immunoglobulin-like receptors (KIRs) which are found on natural killer cells and some T cells; for the CD94:NKG2 family of receptors also expressed mainly by NK cells and for leukocyte immunoglobulin-like receptors (LILRs) on myeloid cells. The aim of this study is to develop an immunogenetic approach for identifying and quantifying the relative contribution of different receptor-ligand interactions to a given HLA class I disease association and then to use this approach to investigate the immune interactions underlying HLA class I disease associations in three viral infections: Human T cell Leukemia Virus type 1, Human Immunodeficiency Virus type 1 and Hepatitis C Virus as well as in the inflammatory condition Crohn's disease.

Broad Recognition of Circulating HIV-1 by HIV-1-Specific Cytotoxic T-Lymphocytes with Strong Ability to Suppress HIV-1 Replication

HIV-1-specific cytotoxic T-lymphocytes (CTLs) with strong abilities to suppress HIV-1 replication and recognize most circulating HIV-1 strains are candidates for effector T cells for cure treatment and prophylactic AIDS vaccine. Previous studies demonstrated that the existence of CTLs specific for 11 epitopes was significantly associated with good clinical outcomes in Japan, although CTLs specific for one of these epitopes select for escape mutations. However, it remains unknown whether the CTLs specific for the remaining 10 epitopes suppress HIV-1 replication in vitro and recognize circulating HIV-1. Here, we investigated the abilities of these CTLs to suppress HIV-1 replication and to recognize variants in circulating HIV-1. CTL clones specific for 10 epitopes had strong abilities to suppress HIV-1 replication in vitro The ex vivo and in vitro analyses of T-cell responses to variant epitope peptides showed that the T cells specific for 10 epitopes recognized mutant peptides which are detected in 84.1% to 98.8% of the circulating HIV-1 strains found in HIV-1-infected Japanese individuals. In addition, the T cells specific for 5 epitopes well recognized target cells infected with 7 mutant viruses that had been detected in >5% of tested individuals. Taken together, these results suggest that CTLs specific for the 10 epitopes effectively suppress HIV-1 replication and broadly recognize the circulating HIV-1 strains in the HIV-1-infected individuals. This study suggests the use of these T cells in clinical trials.IMPORTANCE In recent T-cell AIDS vaccine trials, the vaccines did not prevent HIV-1 infection, although HIV-1-specific T cells were induced in the vaccinated individuals, suggesting that the T cells have a weak ability to suppress HIV-1 replication and fail to recognize circulating HIV-1. We previously demonstrated that the T-cell responses to 10 epitopes were significantly associated with good clinical outcome. However, there is no direct evidence that these T cells have strong abilities to suppress HIV-1 replication and recognize circulating HIV-1. Here, we demonstrated that the T cells specific for the 10 epitopes had strong abilities to suppress HIV-1 replication in vitro Moreover, the T cells cross-recognized most of the circulating HIV-1 in HIV-1-infected individuals. This study suggests the use of T cells specific for these 10 epitopes in clinical trials of T-cell vaccines as a cure treatment.

Genome-scale analysis of evolutionary rate and selection in a fast-expanding Spanish cluster of HIV-1 subtype F1

This work is aimed at assessing the presence of positive selection and/or shifts of the evolutionary rate in a fast-expanding HIV-1 subtype F1 transmission cluster affecting men who have sex with men in Spain. We applied Bayesian coalescent phylogenetics and selection analyses to 23 full-coding region sequences from patients belonging to that cluster, along with other 19 F1 epidemiologically-unrelated sequences. A shift in the overall evolutionary rate of the virus, explained by positively selected sites in the cluster, was detected. We also found one substitution in Nef (H89F) that was specific to the cluster and experienced positive selection. These results suggest that fast transmission could have been facilitated by some inherent genetic properties of this HIV-1 variant.

MuPeXI: prediction of neo-epitopes from tumor sequencing data

Personalization of immunotherapies such as cancer vaccines and adoptive T cell therapy depends on identification of patient-specific neo-epitopes that can be specifically targeted. MuPeXI, the mutant peptide extractor and informer, is a program to identify tumor-specific peptides and assess their potential to be neo-epitopes. The program input is a file with somatic mutation calls, a list of HLA types, and optionally a gene expression profile. The output is a table with all tumor-specific peptides derived from nucleotide substitutions, insertions, and deletions, along with comprehensive annotation, including HLA binding and similarity to normal peptides. The peptides are sorted according to a priority score which is intended to roughly predict immunogenicity. We applied MuPeXI to three tumors for which predicted MHC-binding peptides had been screened for T cell reactivity, and found that MuPeXI was able to prioritize immunogenic peptides with an area under the curve of 0.63. Compared to other available tools, MuPeXI provides more information and is easier to use. MuPeXI is available as stand-alone software and as a web server at http://www.cbs.dtu.dk/services/MuPeXI .

Identification of Immunogenic Cytotoxic T Lymphocyte Epitopes Containing Drug Resistance Mutations in Antiretroviral Treatment-Naïve HIV-Infected Individuals

Background

Therapeutic HIV vaccines may prove helpful to intensify antiretroviral treatment (ART) efficacy and may be an integral part of future cure strategies.

Methods

We examined IFN-gamma ELISpot responses to a panel of 218 HIV clade B consensus-based HIV protease-reverse transcriptase peptides, designed to mimic previously described and predicted cytotoxic T lymphocyte epitopes overlapping drug resistance (DR) positions, that either included the consensus sequence or the DR variant sequence, in 49 ART-naïve HIV-infected individuals. Next generation sequencing was used to assess the presence of minority DR variants in circulating viral populations.

Results

Although a wide spectrum of differential magnitudes of response to DR vs. WT peptide pairs was observed, responses to DR peptides were frequent and strong in the study cohort. No difference between the median magnitudes of response to DR vs. WT peptides was observed. Interestingly, of the 22 peptides that were recognized by >15% of the participants, two-thirds (64%) corresponded to DR peptides. When analysing responses per peptide pair per individual, responses to only WT (median 4 pairs/individual) or DR (median 6 pairs/individual) were more common than responses to both WT and DR (median 2 pairs/individual; p<0.001). While the presence of ELISpot responses to WT peptides was frequently associated with the presence of the corresponding peptide sequence in the patient’s virus (mean 68% of cases), responses to DR peptides were generally not associated with the presence of DR mutations in the viral population, even at low frequencies (mean 1.4% of cases; p = 0.0002).

Conclusions

Our data suggests that DR peptides are frequently immunogenic and raises the potential benefit of broadening the antigens included in a therapeutic vaccine approach to immunogenic epitopes containing common DR sequences. Further studies are needed to assess the quality of responses elicited by DR peptides.

Improved structural method for T-cell cross-reactivity prediction

Cytotoxic T-lymphocytes (CTLs) are the key players of adaptive cellular immunity, being able to identify and eliminate infected cells through the interaction with peptide-loaded major histocompatibility complexes class I (pMHC-I). Despite the high specificity of this interaction, a given lymphocyte is actually able to recognize more than just one pMHC-I complex, a phenomenon referred as cross-reactivity. In the present work we describe the use of pMHC-I structural features as input for multivariate statistical methods, to perform standardized structure-based predictions of cross-reactivity among viral epitopes. Our improved approach was able to successfully identify cross-reactive targets among 28 naturally occurring hepatitis C virus (HCV) variants and among eight epitopes from the four dengue virus serotypes. In both cases, our results were supported by multiscale bootstrap resampling and by data from previously published in vitro experiments. The combined use of data from charges and accessible surface area (ASA) of selected residues over the pMHC-I surface provided a powerful way of assessing the structural features involved in triggering cross-reactive responses. Moreover, the use of an R package (pvclust) for assessing the uncertainty in the hierarchical cluster analysis provided a statistical support for the interpretation of results. Taken together, these methods can be applied to vaccine design, both for the selection of candidates capable of inducing immunity against different targets, or to identify epitopes that could trigger undesired immunological responses.

Comparison of experimental fine-mapping to in silico prediction results of HIV-1 epitopes reveals ongoing need for mapping experiments

Methods for identifying physiologically relevant CD8 T-cell epitopes are critically important not only for the development of T-cell-based vaccines but also for understanding host-pathogen interactions. As experimentally mapping an optimal CD8 T-cell epitope is a tedious procedure, many bioinformatic tools have been developed that predict which peptides bind to a given MHC molecule. We assessed the ability of the CD8 T-cell epitope prediction tools syfpeithi, ctlpred and iedb to foretell nine experimentally mapped optimal HIV-specific epitopes. Randomly - for any of the subjects' HLA type and with any matching score - the optimal epitope was predicted in seven of nine epitopes using syfpeithi, in three of nine epitopes using ctlpred and in all nine of nine epitopes using iedb. The optimal epitope within the three highest ranks was given in four of nine epitopes applying syfpeithi, in two of nine epitopes applying ctlpred and in seven of nine epitopes applying iedb when screening for all of the subjects' HLA types. Knowing the HLA restriction of the peptide of interest improved the ranking of the optimal epitope within the predicted results. Epitopes restricted by common HLA alleles were more likely to be predicted than those restricted by uncommon HLA alleles. Epitopes with aberrant lengths compared with the usual HLA-class I nonamers were most likely not predicted. Application of epitope prediction tools together with literature searches for already described optimal epitopes narrows down the possibilities of optimal epitopes within a screening peptide of interest. However, in our opinion, the actual fine-mapping of a CD8 T-cell epitope cannot yet be replaced.

Functional avidity and IL-2/perforin production is linked to the emergence of mutations within HLA-B*5701-restricted epitopes and HIV-1 disease progression

Viral escape from HIV-1-specific CD8(+) T cells has been demonstrated in numerous studies previously. However, the qualitative features driving the emergence of mutations within epitopes are still unclear. In this study, we aimed to distinguish whether specific functional characteristics of HLA-B*5701-restricted CD8(+) T cells influence the emergence of mutations in high-risk progressors (HRPs) versus low-risk progressors (LRPs). Single-genome sequencing was performed to detect viral mutations (variants) within seven HLA-B*5701-restricted epitopes in Gag (n = 4) and Nef (n = 3) in six untreated HLA-B*5701 subjects followed from early infection up to 7 y. Several well-characterized effector markers (IFN-γ, IL-2, MIP-1β, TNF, CD107a, and perforin) were identified by flow cytometry following autologous (initial and emerging variant/s) epitope stimulations. This study demonstrates that specific functional attributes may facilitate the outgrowth of mutations within HLA-B*5701-restricted epitopes. A significantly lower fraction of IL-2-producing cells and a decrease in functional avidity and polyfunctional sensitivity were evident in emerging epitope variants compared with the initial autologous epitopes. Interestingly, the HRPs mainly drove these differences, whereas the LRPs maintained a directed and maintained functional response against emerging epitope variants. In addition, LRPs induced improved cell-cycle progression and perforin upregulation after autologous and emerging epitope variant stimulations in contrast to HRPs. The maintained quantitative and qualitative features of the CD8(+) T cell responses in LRPs toward emerging epitope variants provide insights into why HLA-B*5701 subjects have different risks of HIV-1 disease progression.

Beyond model antigens: high-dimensional methods for the analysis of antigen-specific T cells

Adaptive immune responses often begin with the formation of a molecular complex between a T-cell receptor (TCR) and a peptide antigen bound to a major histocompatibility complex (MHC) molecule. These complexes are highly variable, however, due to the polymorphism of MHC genes, the random, inexact recombination of TCR gene segments, and the vast array of possible self and pathogen peptide antigens. As a result, it has been very difficult to comprehensively study the TCR repertoire or identify and track more than a few antigen-specific T cells in mice or humans. For mouse studies, this had led to a reliance on model antigens and TCR transgenes. The study of limited human clinical samples, in contrast, requires techniques that can simultaneously survey TCR phenotype and function, and TCR reactivity to many T-cell epitopes. Thanks to recent advances in single-cell and cytometry methodologies, as well as high-throughput sequencing of the TCR repertoire, we now have or will soon have the tools needed to comprehensively analyze T-cell responses in health and disease.

Cross-reactivity analysis of T cell receptors specific for overlapping HIV-1 Nef epitopes of different lengths

Overlapping peptides of different lengths from a certain immunodominant region can be presented by the same HLA class I molecule and elicit different T cell responses. However, how peptide-length specificity of antigen-specific CD8(+) T lymphocytes influence cross-reactivity profiles of these cells remains elusive. This question is particularly important in the face of highly variable pathogens such as HIV-1. Here, we examined this problem by using HLA-B*35:01-restricted CD8(+) T lymphocytes specific for Nef epitopes, i.e., RY11 (RPQVPLRPMTY), VY8 (VPLRPMTY), and RM9 (RPQVPLRPM), in which VY8 and RM9 were contained entirely within RY11, in combination with a T cell receptor (TCR) reconstruction system as well as HLA-B35 tetramers and a set of a single-variant peptide library. The TCR reactivity toward the peptide-length variants was classified into three types: mutually exclusive specificity toward (1) RY11 or (2) VY8 and (3) cross-recognition toward RM9 and RY11. TCR cross-reactivity toward variant peptides was similar within the same peptide-length reactivity type but was markedly different between the types. Thus, TCRs showing similar peptide-length reactivity have shared peptide recognition footprints and thereby similar weakness to antigenic variations, providing us with further insight into the antiviral vaccine design.

Increased sequence coverage through combined targeting of variant and conserved epitopes correlates with control of HIV replication

A major challenge in the development of an HIV vaccine is that of contending with the extensive sequence variability found in circulating viruses. Induction of HIV-specific T-cell responses targeting conserved regions and induction of HIV-specific T-cell responses recognizing a high number of epitope variants have both been proposed as strategies to overcome this challenge. We addressed the ability of cytotoxic T lymphocytes from 30 untreated HIV-infected subjects with and without control of virus replication to recognize all clade B Gag sequence variants encoded by at least 5% of the sequences in the Los Alamos National Laboratory HIV database (1,300 peptides) using gamma interferon and interleukin-2 (IFN-γ/IL-2) FluoroSpot analysis. While targeting of conserved regions was similar in the two groups (P = 0.47), we found that subjects with control of virus replication demonstrated marginally lower recognition of Gag epitope variants than subjects with normal progression (P = 0.05). In viremic controllers and progressors, we found variant recognition to be associated with viral load (r = 0.62, P = 0.001). Interestingly, we show that increased overall sequence coverage, defined as the overall proportion of HIV database sequences targeted through the Gag-specific repertoire, is inversely associated with viral load (r = -0.38, P = 0.03). Furthermore, we found that sequence coverage, but not variant recognition, correlated with increased recognition of a panel of clade B HIV founder viruses (r = 0.50, P = 0.004). We propose sequence coverage by HIV Gag-specific immune responses as a possible correlate of protection that may contribute to control of virus replication. Additionally, sequence coverage serves as a valuable measure by which to evaluate the protective potential of future vaccination strategies.

Properties of MHC class I presented peptides that enhance immunogenicity

T-cells have to recognize peptides presented on MHC molecules to be activated and elicit their effector functions. Several studies demonstrate that some peptides are more immunogenic than others and therefore more likely to be T-cell epitopes. We set out to determine which properties cause such differences in immunogenicity. To this end, we collected and analyzed a large set of data describing the immunogenicity of peptides presented on various MHC-I molecules. Two main conclusions could be drawn from this analysis: First, in line with previous observations, we showed that positions P4–6 of a presented peptide are more important for immunogenicity. Second, some amino acids, especially those with large and aromatic side chains, are associated with immunogenicity. This information was combined into a simple model that was used to demonstrate that immunogenicity is, to a certain extent, predictable. This model (made available at http://tools.iedb.org/immunogenicity/) was validated with data from two independent epitope discovery studies. Interestingly, with this model we could show that T-cells are equipped to better recognize viral than human (self) peptides. After the past successful elucidation of different steps in the MHC-I presentation pathway, the identification of variables that influence immunogenicity will be an important next step in the investigation of T-cell epitopes and our understanding of cellular immune responses.

T-cells have to recognize peptides presented on MHC molecules to be activated and elicit their effector functions. Some peptide-MHC-I complexes (pMHCs) are better recognized by T-cells; we call such pMHCs more immunogenic. For other pMHCs, no recognizing T-cells seem to exist; we call such pMHCs non-immunogenic. We set out to determine which properties of pMHCs cause such differences in immunogenicity, by carefully collecting a large set of immunogenic and non-immunogenic pMHCs, and analysing the difference between these sets. Two important observations were made: First, in line with previous observations, we showed that positions P4–6 of a presented peptide are more important for immunogenicity. Second, some amino acids, especially those with large and aromatic side chains, seem to be better recognized by T-cells as they associate with immunogenicity. Next, this information was combined into a simple model to predict the immunogenicity of new pMHCs (this model is made available at http://tools.iedb.org/immunogenicity/). Interestingly, with this model we could show that T-cells are equipped to strongly recognize viral peptides. After the past successful elucidation of different steps in the MHC-I presentation pathway, the identification of variables that influence immunogenicity will be an important next step in the investigation of T-cell epitopes and our understanding of cellular immune responses.

Peptide-based vaccinology: experimental and computational approaches to target hypervariable viruses through the fine characterization of protective epitopes recognized by monoclonal antibodies and the identification of T-cell-activating peptides

Defining immunogenic domains of viral proteins capable of eliciting a protective immune response is crucial in the development of novel epitope-based prophylactic strategies. This is particularly important for the selective targeting of conserved regions shared among hypervariable viruses. Studying postinfection and postimmunization sera, as well as cloning and characterization of monoclonal antibodies (mAbs), still represents the best approach to identify protective epitopes. In particular, a protective mAb directed against conserved regions can play a key role in immunogen design and in human therapy as well. Experimental approaches aiming to characterize protective mAb epitopes or to identify T-cell-activating peptides are often burdened by technical limitations and can require long time to be correctly addressed. Thus, in the last decade many epitope predictive algorithms have been developed. These algorithms are continually evolving, and their use to address the empirical research is widely increasing. Here, we review several strategies based on experimental techniques alone or addressed by in silico analysis that are frequently used to predict immunogens to be included in novel epitope-based vaccine approaches. We will list the main strategies aiming to design a new vaccine preparation conferring the protection of a neutralizing mAb combined with an effective cell-mediated response.

CD8(+) T cell cross-reactivity profiles and HIV-1 immune escape towards an HLA-B35-restricted immunodominant Nef epitope

Antigen cross-reactivity is an inbuilt feature of the T cell compartment. However, little is known about the flexibility of T cell recognition in the context of genetically variable pathogens such as HIV-1. In this study, we used a combinatorial library containing 24 billion octamer peptides to characterize the cross-reactivity profiles of CD8⁺ T cells specific for the immunodominant HIV-1 subtype B Nef epitope VY8 (VPLRPMTY) presented by HLA-B^*35∶01. In conjunction, we examined naturally occurring antigenic variations within the VY8 epitope. Sequence analysis of plasma viral RNA isolated from 336 HIV-1-infected individuals revealed variability at position (P) 3 and P8 of VY8; Phe at P8, but not Val at P3, was identified as an HLA-B^*35∶01-associated polymorphism. VY8-specific T cells generated from several different HIV-1-infected patients showed unique and clonotype-dependent cross-reactivity footprints. Nonetheless, all T cells recognized both the index Leu and mutant Val at P3 equally well. In contrast, competitive titration assays revealed that the Tyr to Phe substitution at P8 reduced T cell recognition by 50–130 fold despite intact peptide binding to HLA-B^*35∶01. These findings explain the preferential selection of Phe at the C-terminus of VY8 in HLA-B^*35∶01⁺ individuals and demonstrate that HIV-1 can exploit the limitations of T cell recognition in vivo.

Targeting of conserved gag-epitopes in early HIV infection is associated with lower plasma viral load and slower CD4(+) T cell depletion

We aimed to investigate whether the character of the immunodominant HIV-Gag peptide (variable or conserved) targeted by CD8(+) T cells in early HIV infection would influence the quality and quantity of T cell responses, and whether this would affect the rate of disease progression. Treatment-naive HIV-infected study subjects within the OPTIONS cohort at the University of California, San Francisco, were monitored from an estimated 44 days postinfection for up to 6 years. CD8(+) T cells responses targeting HLA-matched HIV-Gag-epitopes were identified and characterized by multicolor flow cytometry. The autologous HIV gag sequences were obtained. We demonstrate that patients targeting a conserved HIV-Gag-epitope in early infection maintained their epitope-specific CD8(+) T cell response throughout the study period. Patients targeting a variable epitope showed decreased immune responses over time, although there was no limitation of the functional profile, and they were likely to target additional variable epitopes. Maintained immune responses to conserved epitopes were associated with no or limited sequence evolution within the targeted epitope. Patients with immune responses targeting conserved epitopes had a significantly lower median viral load over time compared to patients with responses targeting a variable epitope (0.63 log(10) difference). Furthermore, the rate of CD4(+) T cell decline was slower for subjects targeting a conserved epitope (0.85% per month) compared to subjects targeting a variable epitope (1.85% per month). Previous studies have shown that targeting of antigens based on specific HLA types is associated with a better disease course. In this study we show that categorizing epitopes based on their variability is associated with clinical outcome.

Characterization of HIV-specific CD4+ T cell responses against peptides selected with broad population and pathogen coverage

CD4+ T cells orchestrate immunity against viral infections, but their importance in HIV infection remains controversial. Nevertheless, comprehensive studies have associated increase in breadth and functional characteristics of HIV-specific CD4+ T cells with decreased viral load. A major challenge for the identification of HIV-specific CD4+ T cells targeting broadly reactive epitopes in populations with diverse ethnic background stems from the vast genomic variation of HIV and the diversity of the host cellular immune system. Here, we describe a novel epitope selection strategy, PopCover, that aims to resolve this challenge, and identify a set of potential HLA class II-restricted HIV epitopes that in concert will provide optimal viral and host coverage. Using this selection strategy, we identified 64 putative epitopes (peptides) located in the Gag, Nef, Env, Pol and Tat protein regions of HIV. In total, 73% of the predicted peptides were found to induce HIV-specific CD4+ T cell responses. The Gag and Nef peptides induced most responses. The vast majority of the peptides (93%) had predicted restriction to the patient’s HLA alleles. Interestingly, the viral load in viremic patients was inversely correlated to the number of targeted Gag peptides. In addition, the predicted Gag peptides were found to induce broader polyfunctional CD4+ T cell responses compared to the commonly used Gag-p55 peptide pool. These results demonstrate the power of the PopCover method for the identification of broadly recognized HLA class II-restricted epitopes. All together, selection strategies, such as PopCover, might with success be used for the evaluation of antigen-specific CD4+ T cell responses and design of future vaccines.

Predictions versus high-throughput experiments in T-cell epitope discovery: competition or synergy?

Prediction methods as well as experimental methods for T-cell epitope discovery have developed significantly in recent years. High-throughput experimental methods have made it possible to perform full-length protein scans for epitopes restricted to a limited number of MHC alleles. The high costs and limitations regarding the number of proteins and MHC alleles that are feasibly handled by such experimental methods have made in silico prediction models of high interest. MHC binding prediction methods are today of a very high quality and can predict MHC binding peptides with high accuracy. This is possible for a large range of MHC alleles and relevant length of binding peptides. The predictions can easily be performed for complete proteomes of any size. Prediction methods are still, however, dependent on good experimental methods for validation, and should merely be used as a guide for rational epitope discovery. We expect prediction methods as well as experimental validation methods to continue to develop and that we will soon see clinical trials of products whose development has been guided by prediction methods.

Degenerate T-cell recognition of peptides on MHC molecules creates large holes in the T-cell repertoire

The cellular immune system screens peptides presented by host cells on MHC molecules to assess if the cells are infected. In this study we examined whether the presented peptides contain enough information for a proper self/nonself assessment by comparing the presented human (self) and bacterial or viral (nonself) peptides on a large number of MHC molecules. For all MHC molecules tested, only a small fraction of the presented nonself peptides from 174 species of bacteria and 1000 viral proteomes ([Formula: see text]0.2%) is shown to be identical to a presented self peptide. Next, we use available data on T-cell receptor-peptide-MHC interactions to estimate how well T-cells distinguish between similar peptides. The recognition of a peptide-MHC by the T-cell receptor is flexible, and as a result, about one-third of the presented nonself peptides is expected to be indistinguishable (by T-cells) from presented self peptides. This suggests that T-cells are expected to remain tolerant for a large fraction of the presented nonself peptides, which provides an explanation for the "holes in the T-cell repertoire" that are found for a large fraction of foreign epitopes. Additionally, this overlap with self increases the need for efficient self tolerance, as many self-similar nonself peptides could initiate an autoimmune response. Degenerate recognition of peptide-MHC-I complexes by T-cells thus creates large and potentially dangerous overlaps between self and nonself.

Prediction of epitopes using neural network based methods

In this paper, we describe the methodologies behind three different aspects of the NetMHC family for prediction of MHC class I binding, mainly to HLAs. We have updated the prediction servers, NetMHC-3.2, NetMHCpan-2.2, and a new consensus method, NetMHCcons, which, in their previous versions, have been evaluated to be among the very best performing MHC:peptide binding predictors available. Here we describe the background for these methods, and the rationale behind the different optimization steps implemented in the methods. We go through the practical use of the methods, which are publicly available in the form of relatively fast and simple web interfaces. Furthermore, we will review results obtained in actual epitope discovery projects where previous implementations of the described methods have been used in the initial selection of potential epitopes. Selected potential epitopes were all evaluated experimentally using ex vivo assays.

NetMHCcons: a consensus method for the major histocompatibility complex class I predictions

A key role in cell-mediated immunity is dedicated to the major histocompatibility complex (MHC) molecules that bind peptides for presentation on the cell surface. Several in silico methods capable of predicting peptide binding to MHC class I have been developed. The accuracy of these methods depends on the data available characterizing the binding specificity of the MHC molecules. It has, moreover, been demonstrated that consensus methods defined as combinations of two or more different methods led to improved prediction accuracy. This plethora of methods makes it very difficult for the non-expert user to choose the most suitable method for predicting binding to a given MHC molecule. In this study, we have therefore made an in-depth analysis of combinations of three state-of-the-art MHC-peptide binding prediction methods (NetMHC, NetMHCpan and PickPocket). We demonstrate that a simple combination of NetMHC and NetMHCpan gives the highest performance when the allele in question is included in the training and is characterized by at least 50 data points with at least ten binders. Otherwise, NetMHCpan is the best predictor. When an allele has not been characterized, the performance depends on the distance to the training data. NetMHCpan has the highest performance when close neighbours are present in the training set, while the combination of NetMHCpan and PickPocket outperforms either of the two methods for alleles with more remote neighbours. The final method, NetMHCcons, is publicly available at www.cbs.dtu.dk/services/NetMHCcons , and allows the user in an automatic manner to obtain the most accurate predictions for any given MHC molecule.

Translation of HLA-HIV associations to the cellular level: HIV adapts to inflate CD8 T cell responses against Nef and HLA-adapted variant epitopes

Strong statistical associations between polymorphisms in HIV-1 population sequences and carriage of HLA class I alleles have been widely used to identify possible sites of CD8 T cell immune selection in vivo. However, there have been few attempts to prospectively and systematically test these genetic hypotheses arising from population-based studies at a cellular, functional level. We assayed CD8 T cell epitope-specific IFN-γ responses in 290 individuals from the same cohort, which gave rise to 874 HLA-HIV associations in genetic analyses, taking into account autologous viral sequences and individual HLA genotypes. We found immunological evidence for 58% of 374 associations tested as sites of primary immune selection and identified up to 50 novel HIV-1 epitopes using this reverse-genomics approach. Many HLA-adapted epitopes elicited equivalent or higher-magnitude IFN-γ responses than did the nonadapted epitopes, particularly in Nef. At a population level, inclusion of all of the immunoreactive variant CD8 T cell epitopes in Gag, Pol, Nef, and Env suggested that HIV adaptation leads to an inflation of Nef-directed immune responses relative to other proteins. We concluded that HLA-HIV associations mark viral epitopes subject to CD8 T cell selection. These results can be used to guide functional studies of specific epitopes and escape mutations, as well as to test, train, and evaluate analytical models of viral escape and fitness. The inflation of Nef and HLA-adapted variant responses may have negative effects on natural and vaccine immunity against HIV and, therefore, has implications for diversity coverage approaches in HIV vaccine design.

T-cell immune responses against Env from CRF12_BF and subtype B HIV-1 show high clade-specificity that can be overridden by multiclade immunizations

Background

The extreme genetic diversity of the human immunodeficiency virus type 1 (HIV-1) poses a daunting challenge to the generation of an effective AIDS vaccine. In Argentina, the epidemic is characterized by the high prevalence of infections caused by subtype B and BF variants. The aim of this study was to characterize in mice the immunogenic and antigenic properties of the Env protein from CRF12_BF in comparison with clade B, employing prime-boost schemes with the combination of recombinant DNA and vaccinia virus (VV) vectors.

Methodology/Principal Findings

As determined by ELISPOT from splenocytes of animals immunized with either EnvBF or EnvB antigens, the majority of the cellular responses to Env were found to be clade-specific. A detailed peptide mapping of the responses reveal that when there is cross-reactivity, there are no amino acid changes in the peptide sequence or were minimal and located at the peptide ends. In those cases, analysis of T cell polifunctionality and affinity indicated no differences with respect to the cellular responses found against the original homologous sequence.

Significantly, application of a mixed immunization combining both clades (B and BF) induced a broader cellular response, in which the majority of the peptides targeted after the single clade vaccinations generated a positive response. In this group we could also find significant cellular and humoral responses against the whole gp120 protein from subtype B.

Conclusions/Significance

This work has characterized for the first time the immunogenic peptides of certain EnvBF regions, involved in T cell responses. It provides evidence that to improve immune responses to HIV there is a need to combine Env antigens from different clades, highlighting the convenience of the inclusion of BF antigens in future vaccines for geographic regions where these HIV variants circulate.

State of the art and challenges in sequence based T-cell epitope prediction

Sequence based T-cell epitope predictions have improved immensely in the last decade. From predictions of peptide binding to major histocompatibility complex molecules with moderate accuracy, limited allele coverage, and no good estimates of the other events in the antigen-processing pathway, the field has evolved significantly. Methods have now been developed that produce highly accurate binding predictions for many alleles and integrate both proteasomal cleavage and transport events. Moreover have so-called pan-specific methods been developed, which allow for prediction of peptide binding to MHC alleles characterized by limited or no peptide binding data. Most of the developed methods are publicly available, and have proven to be very useful as a shortcut in epitope discovery. Here, we will go through some of the history of sequence-based predictions of helper as well as cytotoxic T cell epitopes. We will focus on some of the most accurate methods and their basic background.

Identification of CD8+ T cell epitopes in the West Nile virus polyprotein by reverse-immunology using NetCTL

Background

West Nile virus (WNV) is a growing threat to public health and a greater understanding of the immune response raised against WNV is important for the development of prophylactic and therapeutic strategies.

Methodology/Principal Findings

In a reverse-immunology approach, we used bioinformatics methods to predict WNV-specific CD8⁺ T cell epitopes and selected a set of peptides that constitutes maximum coverage of 20 fully-sequenced WNV strains. We then tested these putative epitopes for cellular reactivity in a cohort of WNV-infected patients. We identified 26 new CD8⁺ T cell epitopes, which we propose are restricted by 11 different HLA class I alleles. Aiming for optimal coverage of human populations, we suggest that 11 of these new WNV epitopes would be sufficient to cover from 48% to 93% of ethnic populations in various areas of the World.

Conclusions/Significance

The 26 identified CD8⁺ T cell epitopes contribute to our knowledge of the immune response against WNV infection and greatly extend the list of known WNV CD8⁺ T cell epitopes. A polytope incorporating these and other epitopes could possibly serve as the basis for a WNV vaccine.