peptide-MHC Binding Research Articles

Energy landscapes of peptide-MHC binding.

Molecules of the Major Histocompatibility Complex (MHC) present short protein fragments on the cell surface, an important step in T cell immune recognition. MHC-I molecules process peptides from intracellular proteins; MHC-II molecules act in antigen-presenting cells and present peptides derived from extracellular proteins. Here we show that the sequence-dependent energy landscapes of MHC-peptide binding encode class-specific nonlinearities (epistasis). MHC-I has a smooth landscape with global epistasis; the binding energy is a simple deformation of an underlying linear trait. This form of epistasis enhances the discrimination between strong-binding peptides. In contrast, MHC-II has a rugged landscape with idiosyncratic epistasis: binding depends on detailed amino acid combinations at multiple positions of the peptide sequence. The form of epistasis affects the learning of energy landscapes from training data. For MHC-I, a low-complexity problem, we derive a simple matrix model of binding energies that outperforms current models trained by machine learning. For MHC-II, higher complexity prevents learning by simple regression methods. Epistasis also affects the energy and fitness effects of mutations in antigen-derived peptides (epitopes). In MHC-I, large-effect mutations occur predominantly in anchor positions of strong-binding epitopes. In MHC-II, large effects depend on the background epitope sequence but are broadly distributed over the epitope, generating a bigger target for escape mutations due to loss of presentation. Together, our analysis shows how an energy landscape of protein-protein binding constrains the target of escape mutations from T cell immunity, linking the complexity of the molecular interactions to the dynamics of adaptive immune response.

Novel Methodology for the Design of Personalized Cancer Vaccine Targeting Neoantigens: Application to Pancreatic Ductal Adenocarcinoma.

Personalized cancer vaccines have emerged as a promising avenue for cancer treatment or prevention strategies. This approach targets the specific genetic alterations in individual patient's tumors, offering a more personalized and effective treatment option. Previous studies have shown that generalized peptide vaccines targeting a limited scope of gene mutations were ineffective, emphasizing the need for personalized approaches. While studies have explored personalized mRNA vaccines, personalized peptide vaccines have not yet been studied in this context. Pancreatic ductal adenocarcinoma (PDAC) remains challenging in oncology, necessitating innovative therapeutic strategies. In this study, we developed a personalized peptide vaccine design methodology, employing RNA sequencing (RNAseq) to identify prevalent gene mutations underlying PDAC development in a patient solid tumor tissue. We performed RNAseq analysis for trimming adapters, read alignment, and somatic variant calling. We also developed a Python program called SCGeneID, which validates the alignment of the RNAseq analysis. The Python program is freely available to download. Using chromosome number and locus data, SCGeneID identifies the target gene along the UCSC hg38 reference set. Based on the gene mutation data, we developed a personalized PDAC cancer vaccine that targeted 100 highly prevalent gene mutations in two patients. We predicted peptide-MHC binding affinity, immunogenicity, antigenicity, allergenicity, and toxicity for each epitope. Then, we selected the top 50 and 100 epitopes based on our previously published vaccine design methodology. Finally, we generated pMHC-TCR 3D molecular model complex structures, which are freely available to download. The designed personalized cancer vaccine contains epitopes commonly found in PDAC solid tumor tissue. Our personalized vaccine was composed of neoantigens, allowing for a more precise and targeted immune response against cancer cells. Additionally, we identified mutated genes, which were also found in the reference study, where we obtained the sequencing data, thus validating our vaccine design methodology. This is the first study designing a personalized peptide cancer vaccine targeting neoantigens using human patient data to identify gene mutations associated with the specific tumor of interest.

Generating reliable custom MHC tetramers with the QuickSwitchTM tetramer platform.

Abstract Progress in immunology research depends on the continuous development of novel, more adapted and more user-friendly research tools. In 1996 Altman and Davis invented MHC class I tetramers for directly monitoring MHC/peptide specific CD8+ T cells by flow cytometry. Production of MHC tetramers is long and labor intensive and requires advance knowledge of the peptide sequences to be incorporated in these complexes. The recent Covid-19 pandemic has led to new vaccination strategies, which rely on selecting a collection of immunogenic peptides and generating mRNA constructs that contain these sequences. Peptide selection is primarily based on in silico MHC binding tools, which do not always give reliable results. After this first selection step it is necessary to validate MHC peptide binding. For this purpose, a dual use platform was devised, for both assessing MHC peptide binding in a semi-quantitative manner and for T cell staining. The assay is based on a placeholder peptide (or exiting peptide) that can be exchanged for higher affinity binding peptides. Kits with most frequent MHC class I or II alleles are available. Multiple MHC class I and II tetramers can quickly be generated and used for staining within a couple of days.

Targeted acidosis mediated delivery of antigenic MHC-binding peptides.

Cytotoxic T lymphocytes are the primary effector immune cells responsible for protection against cancer, as they target peptide neoantigens presented through the major histocompatibility complex (MHC) on cancer cells, leading to cell death. Targeting peptide-MHC (pMHC) complex offers a promising strategy for immunotherapy due to their specificity and effectiveness against cancer. In this work, we exploit the acidic tumor micro-environment to selectively deliver antigenic peptides to cancer using pH(low) insertion peptides (pHLIP). We demonstrated the delivery of MHC binding peptides directly to the cytoplasm of melanoma cells resulted in the presentation of antigenic peptides on MHC, and activation of T cells. This work highlights the potential of pHLIP as a vehicle for the targeted delivery of antigenic peptides and its presentation via MHC-bound complexes on cancer cell surface for activation of T cells with implications for enhancing anti-cancer immunotherapy.

MORITS: An improved method to predict peptides from heterologous proteins that are recognized by the same T-cell receptor

Antigen-specific priming of T cells results in the activation of T cells that exert effector functions by interaction of their T-cell receptor (TCR) with the corresponding self-MHC molecule presenting a peptide on the surface of a target cell. Such antigen-specific T cells potentially can also interact with peptide-MHC complexes that contain peptides from unrelated antigens, a phenomenon that often is referred to as heterologous immunity. For example, some individuals that were pre-immunized against an allergen, could subsequently mount better anti-viral T-cell responses than non-allergic individuals. So far only few peptide pairs that experimentally have been shown to provoke heterologous immunity were identified, and available prediction tools that can identify potential candidates are imprecise. We developed the MORITS algorithm to rapidly screen large lists of peptides for sequence similarities, while giving enhanced consideration to peptide residues presented by MHC that are particularly relevant for TCR interactions. In combination with established peptide-MHC binding prediction tools, the MORITS algorithm revealed peptide similarities between the SARS-CoV-2 proteome and certain allergens. The method outperformed previously published workflows and may help to identify novel pairs of peptides that mediate heterologous immune responses.

The electrostatic landscape of MHC-peptide binding revealed using inception networks

Predictive modeling of macromolecular recognition and protein-protein complementarity represents one of the cornerstones of biophysical sciences. However, such models are often hindered by the combinatorial complexity of interactions at the molecular interfaces. Exemplary of this problem is peptide presentation by the highly polymorphic major histocompatibility complex class I (MHC-I) molecule, a principal component of immune recognition. We developed human leukocyte antigen (HLA)-Inception, a deep biophysical convolutional neural network, which integrates molecular electrostatics to capture non-bonded interactions for predicting peptide binding motifs across 5,821 MHC-I alleles. These predictions of generated motifs correlate strongly with experimental peptide binding and presentation data. Beyond molecular interactions, the study demonstrates the application of predicted motifs in analyzing MHC-I allele associations with HIV disease progression and patient response to immune checkpoint inhibitors. A record of this paper's transparent peer review process is included in the supplemental information.

Graph-pMHC: graph neural network approach to MHC class II peptide presentation and antibody immunogenicity.

Antigen presentation on MHC class II (pMHCII presentation) plays an essential role in the adaptive immune response to extracellular pathogens and cancerous cells. But it can also reduce the efficacy of large-molecule drugs by triggering an anti-drug response. Significant progress has been made in pMHCII presentation modeling due to the collection of large-scale pMHC mass spectrometry datasets (ligandomes) and advances in machine learning. Here, we develop graph-pMHC, a graph neural network approach to predict pMHCII presentation. We derive adjacency matrices for pMHCII using Alphafold2-multimer and address the peptide-MHC binding groove alignment problem with a simple graph enumeration strategy. We demonstrate that graph-pMHC dramatically outperforms methods with suboptimal inductive biases, such as the multilayer-perceptron-based NetMHCIIpan-4.0 (+20.17% absolute average precision). Finally, we create an antibody drug immunogenicity dataset from clinical trial data and develop a method for measuring anti-antibody immunogenicity risk using pMHCII presentation models. Our model increases receiver operating characteristic curve (ROC)-area under the ROC curve (AUC) by 2.57% compared to just filtering peptides by hits in OASis alone for predicting antibody drug immunogenicity.

Abstract 24: Utilizing T cell receptor-based therapy to treat anaplastic thyroid cancer

Abstract Anaplastic thyroid cancer (ATC) is one of the most aggressive solid tumor in thyroid tissue, with a 6-month medium overall survival. The effectiveness of conventional treatments in extending the survival of patients is limited. Therefore, there is an urgent need to develop novel systemic therapeutic interventions. Adoptive cell therapy has revolutionized cancer immunotherapy by successfully treating B-cell malignancies through chimeric antigen receptor-based approaches. For solid tumors, there is a need to identify a broader spectrum of tumor target antigens, and the field has now shifted towards leveraging T cell receptors (TCRs) which can recognize both intracellular and extracellular antigen-derived epitopes. Our team has established an immunogenic epitope discovery pipeline, followed by a specialized protocol for empiric validation of candidate epitopes. We characterized immunogenic epitopes by coupling tandem mass spectrometry of MHC-eluted peptides with tumor proteome and MHC-peptide binding prediction databases. By integrating multi-omics data (genomics, transcriptomics, and immunopeptidomics) from both in-house and publicly available ATC databases, we find that ATC exhibits significant upregulation of pathways associated with microtubule-based movement, chromosome segregation, and kinetochore complex components. We have identified several HLA-A2- and A24-restricted immunogenic peptides representing these pathway-associated proteins that covered more than 60% of the patient population. These immunogenic peptides were subsequently utilized to isolate antigen-specific T cells from peripheral blood and identify cognate TCR. These peptide-specific T lymphocytes elicit cytotoxicity toward multiple HLA-matched ATC cell lines and tumor samples in vitro and in vivo. In conclusion, our studies contribute to the development of a new treatment paradigm for treating ATC. Citation Format: Shao-Hsi Hung, Ying C. Henderson, Pan Ke, Yulun Chiu, Shailbala Singh, Yunyun Chen, Stephen Y. Lai, Cassian Yee. Utilizing T cell receptor-based therapy to treat anaplastic thyroid cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 24.

Abstract 4107: Evaluating the efficacy of personalized neoantigen cancer vaccine in preclinical model: A promising step towards precision immunotherapy

Abstract Background: In the past decade, immunotherapy has transformed cancer treatment, with the development of immune-checkpoint inhibitors and adoptive cell therapies. A novel frontier is personalized vaccines targeting neoantigens unique to tumor cells. Vaccines based on neoantigens have several advantages. First, neoantigens are exclusively expressed by tumor cells and can, therefore, elicit truly tumor-specific T cell responses, thereby preventing ‘off-target’ damage to nonmalignant tissues. Second, neoantigens are de novo epitopes derived from somatic mutations, which presents the possibility to circumvent T cell central tolerance of self-epitopes and thus induce immune responses to tumors. One of the major challenges is the optimal neoantigen discovery/selection. Our study combines in silico HLA binding models with the 'MHC-PepSeq' platform to enhance neoantigen prediction, creating a personalized neoantigen peptide vaccine. We aim to demonstrate its efficacy in a mouse CT26 colon cancer model. Methods: Our method involved whole exome sequencing of tumor and normal DNA to identify neoantigens expressed in CT26 tumors, generating a peptide library. The peptides were further filtered using an in-silico prediction algorithm and our proprietary peptide-MHC binding assay (PepSeq assay), to determine those exhibiting the highest binding affinity with mice HLAs. Vaccines were prepared by dissolving these peptides in sterile PBS. Mice (BALB/C) were subcutaneously implanted with CT26 tumors, then immunized with the vaccine via IV injection in combination with a TLR7/8 agonist. Mice received a second vaccine dose on Day 14, and some groups were treated with an anti-PD-1 agent. Tumors and spleens from the mice were collected, processed into single-cell suspensions, and analyzed by single-cell sequencing, TCR analysis, IHC, and flow cytometry assays. Peripheral blood mononuclear cells (PBMCs) were collected and utilized in MHC tetramer assays. Results: Preliminary in vivo studies demonstrated significant tumor growth restrictions (TGI around 70%) and prolonged tumor-free survival in some mice. Our data showed the generation of neoantigen-specific T cells in vaccinated mice. Remarkably, vaccinated, tumor-free mice, when re-challenged with the same tumor cells, maintained their tumor-free status, suggesting the vaccine induced post-treatment immunological memory. Combining the vaccine with checkpoint inhibition resulted in better tumor growth restriction in mice (TGI around 93%). Conclusion: Overall, our data suggests that our personalized neoantigen vaccine platform has the potential to evoke robust and durable immune responses and, when paired with suitable complementary therapies and immune adjuvants, may serve as an effective cancer treatment. Citation Format: Tithi Ghosh Halder, Kate Gutowsky, Serina Ng, Trason Thode, Alexis Weston, Erin Kelley, Jorge Soria-Bustos, Jorge A. Giron, Taylor Bargenquast, Mohan Kaadige, Sanjana Tripuraneni, Brian Durbin, Shelby Rheinschmidt, Sydney Adamson, Michael Gordon, Justin Moser, John Altin, Raffaella Soldi, Sunil Sharma. Evaluating the efficacy of personalized neoantigen cancer vaccine in preclinical model: A promising step towards precision immunotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4107.

T-cell virtuosity in ''knowing thyself".

Major Histocompatibility Complex (MHC) I and II and the αβ T-cell antigen receptor (TCRαβ) govern fundamental traits of adaptive immunity. They form a membrane-borne ligand-receptor system weighing host proteome integrity to detect contamination by nonself proteins. MHC-I and -II exhibit the "MHC-fold", which is able to bind a large assortment of short peptides as proxies for self and nonself proteins. The ensuing varying surfaces are mandatory ligands for Ig-like TCRαβ highly mutable binding sites. Conserved molecular signatures guide TCRαβ ligand binding sites to focus on the MHC-fold (MHC-restriction) while leaving many opportunities for its most hypervariable determinants to contact the peptide. This riveting molecular strategy affords many options for binding energy compatible with specific recognition and signalling aimed to eradicated microbial pathogens and cancer cells. While the molecular foundations of αβ T-cell adaptive immunity are largely understood, uncertainty persists on how peptide-MHC binding induces the TCRαβ signals that instruct cell-fate decisions. Solving this mystery is another milestone for understanding αβ T-cells' self/nonself discrimination. Recent developments revealing the innermost links between TCRαβ structural dynamics and signalling modality should help dissipate this long-sought-after enigma.

RPEMHC: improved prediction of MHC-peptide binding affinity by a deep learning approach based on residue-residue pair encoding.

Binding of peptides to major histocompatibility complex (MHC) molecules plays a crucial role in triggering T cell recognition mechanisms essential for immune response. Accurate prediction of MHC-peptide binding is vital for the development of cancer therapeutic vaccines. While recent deep learning-based methods have achieved significant performance in predicting MHC-peptide binding affinity, most of them separately encode MHC molecules and peptides as inputs, potentially overlooking critical interaction information between the two. In this work, we propose RPEMHC, a new deep learning approach based on residue-residue pair encoding to predict the binding affinity between peptides and MHC, which encode an MHC molecule and a peptide as a residue-residue pair map. We evaluate the performance of RPEMHC on various MHC-II-related datasets for MHC-peptide binding prediction, demonstrating that RPEMHC achieves better or comparable performance against other state-of-the-art baselines. Moreover, we further construct experiments on MHC-I-related datasets, and experimental results demonstrate that our method can work on both two MHC classes. These extensive validations have manifested that RPEMHC is an effective tool for studying MHC-peptide interactions and can potentially facilitate the vaccine development. The source code of the method along with trained models is freely available at https://github.com/lennylv/RPEMHC.

Designing High Binding Affinity Peptides for MHC Class I Using MAM: An In Silico Approach.

The availability of extensive MHC-peptide binding data has boosted machine learning-based approaches for predicting binding affinity and identifying binding motifs. These computational tools leverage the wealth of binding data to extract essential features and generate a multitude of potential peptides, thereby significantly reducing the cost and time required for experimental procedures. MAM is one such tool for predicting the MHC-I-peptide binding affinity, extracting binding motifs, and generating new peptides with high affinity. This manuscript provides step-by-step guidance on installing, configuring, and executing MAM while also discussing the best practices when using this tool.

The molecular mechanisms of CD8+ T cell responses to SARS-CoV-2 infection mediated by TCR-pMHC interactions.

Cytotoxic CD8+ T lymphocytes (CTLs) have been implicated in the severity of COVID-19. The TCR-pMHC ternary complex, formed by the T cell receptor (TCR) and peptide-MHC (major histocompatibility complex), constitutes the molecular basis of CTL responses against SARS-CoV-2. While numerous studies have been conducted on T cell immunity, the molecular mechanisms underlying CTL-mediated immunity against SARS-CoV-2 infection have not been well elaborated. In this review, we described the association between HLA variants and different immune responses to SARS-CoV-2 infection, which may lead to varying COVID-19 outcomes. We also summarized the specific TCR repertoires triggered by certain SARS-CoV-2 CTL epitopes, which might explain the variations in disease outcomes among different patients. Importantly, we have highlighted the primary strategies used by SARS-CoV-2 variants to evade T-cell killing: disrupting peptide-MHC binding, TCR recognition, and antigen processing. This review provides valuable insights into the molecule mechanism of CTL responses during SARS-CoV-2 infection, aiding efforts to control the pandemic and prepare for future challenges.

MHCSeqNet2-improved peptide-class I MHC binding prediction for alleles with low data.

The binding of a peptide antigen to a Class I major histocompatibility complex (MHC) protein is part of a key process that lets the immune system recognize an infected cell or a cancer cell. This mechanism enabled the development of peptide-based vaccines that can activate the patient's immune response to treat cancers. Hence, the ability of accurately predict peptide-MHC binding is an essential component for prioritizing the best peptides for each patient. However, peptide-MHC binding experimental data for many MHC alleles are still lacking, which limited the accuracy of existing prediction models. In this study, we presented an improved version of MHCSeqNet that utilized sub-word-level peptide features, a 3D structure embedding for MHC alleles, and an expanded training dataset to achieve better generalizability on MHC alleles with small amounts of data. Visualization of MHC allele embeddings confirms that the model was able to group alleles with similar binding specificity, including those with no peptide ligand in the training dataset. Furthermore, an external evaluation suggests that MHCSeqNet2 can improve the prioritization of T cell epitopes for MHC alleles with small amount of training data. The source code and installation instruction for MHCSeqNet2 are available at https://github.com/cmb-chula/MHCSeqNet2.

Accurate modeling of peptide-MHC structures with AlphaFold

Major histocompatibility complex (MHC) proteins present peptides on the cell surface for Tcell surveillance. Reliable in silico prediction of which peptides would be presented and which Tcell receptors would recognize them is an important problem in structural immunology. Here, we introduce an AlphaFold-based pipeline for predicting the three-dimensional structures of peptide-MHC complexes for class I and class II MHC molecules. Our method demonstrates high accuracy, outperforming existing tools in class I modeling accuracy and class II peptide register prediction. We validate its performance and utility with new experimental data on a recently described cancer neoantigen/wild-type peptide pair and explore applications toward improving peptide-MHC binding prediction.

PANDORA v2.0: Benchmarking peptide-MHC II models and software improvements.

T-cell specificity to differentiate between self and non-self relies on T-cell receptor (TCR) recognition of peptides presented by the Major Histocompatibility Complex (MHC). Investigations into the three-dimensional (3D) structures of peptide:MHC (pMHC) complexes have provided valuable insights of MHC functions. Given the limited availability of experimental pMHC structures and considerable diversity of peptides and MHC alleles, it calls for the development of efficient and reliable computational approaches for modeling pMHC structures. Here we present an update of PANDORA and the systematic evaluation of its performance in modelling 3D structures of pMHC class II complexes (pMHC-II), which play a key role in the cancer immune response. PANDORA is a modelling software that can build low-energy models in a few minutes by restraining peptide residues inside the MHC-II binding groove. We benchmarked PANDORA on 136 experimentally determined pMHC-II structures covering 44 unique αβ chain pairs. Our pipeline achieves a median backbone Ligand-Root Mean Squared Deviation (L-RMSD) of 0.42 Å on the binding core and 0.88 Å on the whole peptide for the benchmark dataset. We incorporated software improvements to make PANDORA a pan-allele framework and improved the user interface and software quality. Its computational efficiency allows enriching the wealth of pMHC binding affinity and mass spectrometry data with 3D models. These models can be used as a starting point for molecular dynamics simulations or structure-boosted deep learning algorithms to identify MHC-binding peptides. PANDORA is available as a Python package through Conda or as a source installation at https://github.com/X-lab-3D/PANDORA.

Personal and Shared Tumor-Antigen Prioritization in Diffuse Large B Cell Lymphoma Patients Undergoing CD19 CAR T Cell Treatment

Backgroun d: Durable response to Axi-cel CD19 CAR T cell therapy is associated with activation and clonotypic expansion of the host T cell repertoire (Cheloni et al Blood, ASH, 2022). The characterization of the tumor microenvironment (TME) and the identification of tumor antigens associated with therapeutic efficacy is a critical area of investigation. Aim: We sought to determine the association between the host immune activation in response to tumor-associated and tumor-specific antigens and the clinical response to CAR-T cell treatment in diffuse large B cell lymphoma patients. Methods: RNAseq was performed on FFPE lymphoma samples collected before CAR T cell infusion (N=16). Patients were divided into 3 distinct cohorts: stable progressive disease at 1 month post-CAR T (non-responders, n=2), disease relapse within 6 months post-CAR T (relapsed, n=3), and on-going response lasting greater than 1 year (long-term responders, n=11). For neoantigens, mutations were retrieved following the GATK best practices for RNAseq data and annotated with Variant Effect Predictor (VEP). Tumor-specific antigens and tumor-associated antigens were detected by prioritizing 9-11mer amino acid sequences, preferentially presented by class I molecules, observed solely or primarily in tumor samples, respectively, as compared to a pool of healthy controls derived from 250 RNAseq samples, corresponding to 29 tissues (GTEx Portal). Antigen prioritization was performed following weighted scoring considering RNA expression, translation potential, peptide-MHC binding affinity, predicted immunogenicity, physicochemical properties, and clonality, in the case of neoantigens. Using deconvolution methods and canonical marker expression, we characterized the major immune cell types within the TME. Results: Higher cell proportions of lymphocytes and innate immune cells, including NK, and activated dendritic cells were observed in long-term responders, suggesting that a higher infiltration of these cell types in the tumor bed is associated with a better response to CAR T cell therapy. Notably, a positive correlation between the mutational burden and the a/b clonotypic expansion of peripheral blood cytotoxic T cell populations, detected from single-cell immunoprofiling (Cheloni et al Blood, ASH, 2022) was observed in eight long-term responders, consistent with prior studies highlighting the significance of neoantigens in eliciting robust and durable immune response against cancer cells. A high occurrence of neoantigens originating from the Titin ( TTN) gene in two long-term responders with the highest mutational burden was observed, and from ACTB and ACTG1, mutated genes associated with lymphoid cancers. A substantial enrichment of shared tumor antigens originating from retroelements was observed in long-term responders, suggesting the presence of common epitopes that drive an antigen-specific response. Conclusion: These findings shed light on the potential immunogenicity of specific neoantigens and shared antigens derived from repetitive elements, which could serve as promising targets for immunotherapeutic strategies in lymphoma patients. Understanding the tumor immune landscape and antigenic features associated with long-term response to CAR T cell therapy may pave the way for the development of personalized and effective combinatorial treatments for lymphoid cancers.

Structural basis for T cell recognition of cancer neoantigens and implications for predicting neoepitope immunogenicity.

Adoptive cell therapy (ACT) with tumor-specific T cells has been shown to mediate durable cancer regression. Tumor-specific T cells are also the basis of other therapies, notably cancer vaccines. The main target of tumor-specific T cells are neoantigens resulting from mutations in self-antigens over the course of malignant transformation. The detection of neoantigens presents a major challenge to T cells because of their high structural similarity to self-antigens, and the need to avoid autoimmunity. How different a neoantigen must be from its wild-type parent for it to induce a T cell response is poorly understood. Here we review recent structural and biophysical studies of T cell receptor (TCR) recognition of shared cancer neoantigens derived from oncogenes, including p53R175H, KRASG12D, KRASG12V, HHATp8F, and PIK3CAH1047L. These studies have revealed that, in some cases, the oncogenic mutation improves antigen presentation by strengthening peptide-MHC binding. In other cases, the mutation is detected by direct interactions with TCR, or by energetically driven or other indirect strategies not requiring direct TCR contacts with the mutation. We also review antibodies designed to recognize peptide-MHC on cell surfaces (TCR-mimic antibodies) as an alternative to TCRs for targeting cancer neoantigens. Finally, we review recent computational advances in this area, including efforts to predict neoepitope immunogenicity and how these efforts may be advanced by structural information on peptide-MHC binding and peptide-MHC recognition by TCRs.

Prediction of Graft-Versus-Host Disease (GVHD) in Recipients of Hematopoietic Cell Transplant(alloHCT) from a Single Mismatched Unrelated Donor Using a Highly Multiplexed Proteomics Assay: MHC-Pepseq

Introduction: Graft-versus-host disease (GVHD) remains a major barrier of allogenic hematopoietic stem cell transplantation (HCT). While direct recognition of mismatched HLA molecules is believed to be the major driver of alloimmune T cell response, indirect presentation of mismatched host class I or class II proteins may also contribute to CD4 T cell-mediated GVHD or graft-versus-leukemia (GVL). Using a novel highly multiplexed peptide-MHC binding assay (MHC-PepSeq), we developed a score of HLA-DRB1 restricted and binding allogeneic peptide sequences for each donor/recipient pair in a CIBMTR cohort of mismatched adult HCT recipients with hematological malignancies. We hypothesized that the risk score derived from the MHC-PepSeq assay is associated with the incidence and severity of acute and chronic GVHD. Methods: Using public population allele frequencies (allelefrequencies.net), we identified a set of alleles that covers >95% of HLA genotypes of 3 major US populations (European Caucasian, African American, Mexican Chicano) at 9 HLA-loci (-A, -B, -C, -DRA1, -DRB1, -DQA1, -DQB1, -DPA1, -DPB1) and converted their sequences to 7,744 unique, densely-tiled 15-mer peptides. A highly multiplexed synthesis protocol (“PepSeq”) was used to generate a library of DNA-barcoded peptides of the corresponding sequences, which we used to generate experimental binding measurements of each peptide against 12 common HLA-DRB1 molecules. The resulting data were then used to enumerate DR-binding peptides present in recipient HLA but not donor HLA genotype (“allopeptide score”), for each donor-recipient pair in a cohort of HLA-A, B, C and DRB1 (8/8) matched unrelated donors with a mismatch in HLA-DPB1 and a cohort of 7/8 Class I mismatched unrelated donors. Univariate analysis was used to compare clinical demographics and allopeptide association with clinical outcomes. In multivariable analysis, a logistic regression model was used to analyze aGVHD II-IV at day 100 (primary outcome), and Cox proportional hazards were applied for OS, relapse, chronic GVHD (cGVHD) and non-relapse mortality (NRM) (secondary outcomes). Results: In the 8/8 cohort, 6679 matched unrelated pairs were identified with allopeptide detected in 1563 (30.5%). In the 7/8 cohort, 1133 pairs were identified with allopeptide detected in 156 (15.9%). Patient characteristics were balanced in both cohorts and described in Table 1. In the 8/8 cohort, the cumulative incidence of aGVHD II-IV at day 100 post HCT was 42.3% (95% CI: 40.9-43.7%) in patients with a detectable allopeptide and 45.5% (95%CI: 43.0-48.0%) in patients with non-detectable allopeptide, respectively (p=.069). No significant differences were observed in univariate analysis for cGVHD, NRM, OS. In multivariate analysis there were no significant associations between presence of detectable allopeptide (Y/N) and aGVHD II-IV, cGVHD, OS, and NRM. However, a significant interaction between disease and allopeptide status was detected in the relapse model (P=0.0020), with allopeptide presence associated with a decreased risk of relapse in ALL (HR 0.71; 95%CI:0.54-0.94, p=0.0168) and an increased risk of relapse in AML (HR 1.23; 95%CI:1.05-1.43, p=0.0085). Allopeptide presence was not significantly associated with relapse in MDS (HR 0.89; 95% CI:0.74-1.06 p=0.1901). Adjusted cumulative incidence of relapse is shown in Figure 1. In the 7/8 cohort, cumulative incidence of aGVHD II-IV at day 100 post HCT was 43.4% (95% CI: 40.2-46.5%) and 41.8% (95% CI: 34.1-49.7%) (p=.77) in the allopeptide-detectable and non-detectable groups, respectively. In univariate analysis there were no statistically significant differences noted in cGVHD, NRM, OS. In multivariate analysis allopeptide presence was not significantly associated with any of the endpoints aGVHD II-IV, cGVHD, OS, relapse or NRM. Conclusion: In a large CIBMTR 8/8 and 7/8 matched unrelated donor cohort, allopeptide scores derived from MHC-pepseq were not predictive of aGVHD, possibly due to the limitation of the assay (i.e., low sensitivity, limited to class II presentation - HLA-DRB1 in this analysis) and/or low antigenicity in HLA molecules when indirectly presented as minor histocompatibility antigens. The observed reverse association of allopeptide presence with relapse in AML and ALL requires further investigation, and subgroup analyses for individual HLA-DRB1 presenting allele are currently underway.

Coarse-grained molecular dynamics-guided immunoinformatics to explain the binder and non-binder classification of Cytotoxic T-cell epitope for SARS-CoV-2 peptide-based vaccine discovery.

Epitope-based peptide vaccine can elicit T-cell immunity against SARS-CoV-2 to clear the infection. However, finding the best epitope from the whole antigen is challenging. A peptide screening using immunoinformatics usually starts from MHC-binding peptide, immunogenicity, cross-reactivity with the human proteome, to toxicity analysis. This pipeline classified the peptides into three categories, i.e., strong-, weak-, and non-binder, without incorporating the structural aspect. For this reason, the molecular detail that discriminates the binders from non-binder is interesting to be investigated. In this study, five CTL epitopes against HLA-A*02:01 were identified from the coarse-grained molecular dynamics-guided immunoinformatics screening. The strong binder showed distinctive activities from the non-binder in terms of structural and energetic properties. Furthermore, the second residue from the nonameric peptide was most important in the interaction with HLA-A*02:01. By understanding the nature of MHC-peptide interaction, we hoped to improve the chance of finding the best epitope for a peptide vaccine candidate.