Towards guided mutagenesis: Gaussian process regression predicts MHC class II antigen mutant binding

Mapping Intimacies ◽

10.1101/2021.04.14.439878 ◽

2021 ◽

Author(s):

David R. Bell ◽

Serena H. Chen

Keyword(s):

Gaussian Process ◽

Binding Affinity ◽

Mhc Class Ii ◽

Class Ii ◽

Average Error ◽

Minimal Amount ◽

Binding Affinities ◽

Binding Interaction ◽

Anchor Residue ◽

Auc Value

Antigen-specific immunotherapies (ASI) require successful loading and presentation of antigen peptide into the major histocompatibility complex (MHC) binding cleft. One route of ASI design is to mutate native antigens for either stronger or weaker binding interaction to MHC. Exploring all possible mutations is costly both experimentally and computationally. To reduce experimental and computational expense, here we investigate the minimal amount of prior data required to accurately predict the relative binding affinity of point mutations for peptide-MHC class II (pMHCII) binding. Using data from different residue subsets, we interpolate pMHCII mutant binding affinities by Gaussian process (GP) regression of residue volume and hydrophobicity. We apply GP regression to an experimental dataset from the Immune Epitope Database, and theoretical datasets from NetMHCIIpan and Free Energy Perturbation calculations. We find that GP regression can predict binding affinities of 9 neutral residues from a 6-residue subset with an average R2 coefficient of determination value of 0.62 ± 0.04 (± 95% CI), average error of 0.09 ± 0.01 kcal/mol (± 95% CI), and with an ROC AUC value of 0.92 for binary classification of enhanced or diminished binding affinity. Similarly, metrics increase to an R2 value of 0.69 ± 0.04, average error of 0.07 ± 0.01 kcal/mol, and an ROC AUC value of 0.94 for predicting 7 neutral residues from an 8-residue subset. Our work finds that prediction is most accurate for neutral residues at anchor residue sites without register shift. This work holds relevance to predicting pMHCII binding and accelerating ASI design.

Download Full-text

DeepMHCII: A Novel Binding Core-Aware Deep Interaction Model for Accurate MHC II-peptide Binding Affinity Prediction

10.1101/2021.12.27.474242 ◽

2021 ◽

Author(s):

Ronghui You ◽

Wei Qu ◽

Hiroshi Mamitsuka ◽

Shanfeng Zhu

Keyword(s):

Deep Learning ◽

Binding Affinity ◽

Mhc Class Ii ◽

Large Scale ◽

Peptide Binding ◽

Class Ii ◽

Biological Knowledge ◽

Binding Interaction ◽

Binding Core ◽

Peptide Binding Affinity

Computationally predicting MHC-peptide binding affinity is an important problem in immunological bioinformatics. Recent cutting-edge deep learning-based methods for this problem are unable to achieve satisfactory performance for MHC class II molecules. This is because such methods generate the input by simply concatenating the two given sequences: (the estimated binding core of) a peptide and (the pseudo sequence of) an MHC class II molecule, ignoring the biological knowledge behind the interactions of the two molecules. We thus propose a binding core-aware deep learning-based model, DeepMHCII, with binding interaction convolution layer (BICL), which allows integrating all potential binding cores (in a given peptide) and the MHC pseudo (binding) sequence, through modeling the interaction with multiple convolutional kernels. Extensive empirical experiments with four large-scale datasets demonstrate that DeepMHCII significantly outperformed four state-of-the-art methods under numerous settings, such as five-fold cross-validation, leave one molecule out, validation with independent testing sets, and binding core prediction. All these results with visualization of the predicted binding cores indicate the effectiveness and importance of properly modeling biological facts in deep learning for high performance and knowledge discovery. DeepMHCII is publicly available at https://weilab.sjtu.edu.cn/DeepMHCII/.

Download Full-text

Improved methods for predicting peptide binding affinity to MHC class II molecules

Immunology ◽

10.1111/imm.12889 ◽

2018 ◽

Vol 154 (3) ◽

pp. 394-406 ◽

Cited By ~ 202

Author(s):

Kamilla Kjaergaard Jensen ◽

Massimo Andreatta ◽

Paolo Marcatili ◽

Søren Buus ◽

Jason A. Greenbaum ◽

...

Keyword(s):

Binding Affinity ◽

Mhc Class Ii ◽

Peptide Binding ◽

Class Ii ◽

Improved Methods ◽

Peptide Binding Affinity ◽

Mhc Class Ii Molecules

Download Full-text

A Superantigen-Antibody Fusion Protein for T-Cell Immunotherapy of Human B-Lineage Malignancies

Blood ◽

10.1182/blood.v89.6.2089 ◽

1997 ◽

Vol 89 (6) ◽

pp. 2089-2097 ◽

Cited By ~ 21

Author(s):

Cecilia Gidlöf ◽

Mikael Dohlsten ◽

Peter Lando ◽

Terje Kalland ◽

Christer Sundström ◽

...

Keyword(s):

T Cells ◽

T Cell ◽

Fusion Protein ◽

B Cell ◽

Binding Affinity ◽

Mhc Class Ii ◽

Class Ii ◽

Target Cells ◽

Cell Based Therapy ◽

B Lineage

Abstract The bacterial superantigen staphylococcal enterotoxin A (SEA) is an efficient activator of cytotoxic T cells when presented on major histocompatibility complex (MHC) class II molecules of target cells. Our previous studies showed that such SEA-directed T cells efficiently lysed chronic B-lymphocytic leukemia (B-CLL) cells. Next, we made a mutated SEA–protein A (SEAm-PA) fusion protein with more than 1,000-fold reduced binding affinity for MHC class II compared with native SEA. The fusion protein was successfully used to direct T cells to B-CLL cells coated with different B lineage–directed monoclonal antibodies (MoAbs). In this communication, we constructed a recombinant anti-CD19-Fab-SEAm fusion protein. The MHC class II binding capacity of the SEA part was drastically reduced by a D227A point mutation, whereas the T-cell activation properties were retained. The Fab part of the fusion protein displayed a binding affinity for CD19+ cells in the nanomolar range. The anti-CD19-Fab-SEAm molecule mediated effective, specific, rapid, and perforin-like T-cell lysis of B-CLL cells at low effector to target cell ratios. Normal CD19+ B cells were sensitive to lysis, whereas CD34+ progenitor cells and monocytes/macrophages were resistant. A panel of CD19+ B-cell lines representing different B-cell developmental stages were efficiently lysed, and the sensitivity correlated with surface ICAM-1 expression. The anti-CD19-Fab-SEAm fusion protein mediated highly effective killing of tumor biopsy cells representing several types of B-cell non-Hodgkin's lymphoma (B-NHL). Humanized severe combined immune deficiency (SCID) mice carrying Daudi lymphoma cells were used as an in vivo therapy model for evaluation of the anti-CD19-Fab-SEAm fusion protein. Greater than 90% reduction in tumor weight was recorded in anti-CD19-Fab-SEAm–treated animals compared with control animals receiving an irrelevant Fab-SEAm fusion protein. The present results indicate that MoAb-targeted superantigens (SAgs) may represent a promising approach for T-cell–based therapy of CD19+ B-cell malignancies.

Download Full-text

Structural Basis of Cytochrome c Presentation by IEk

Journal of Experimental Medicine ◽

10.1084/jem.20011971 ◽

2002 ◽

Vol 195 (8) ◽

pp. 1043-1052 ◽

Cited By ~ 38

Author(s):

Daved H. Fremont ◽

Shaodong Dai ◽

Herbert Chiang ◽

Frances Crawford ◽

Philippa Marrack ◽

...

Keyword(s):

T Cell ◽

Cytochrome C ◽

Mhc Class Ii ◽

Class Ii ◽

Structural Basis ◽

Anchor Residue ◽

T Cell Antigens ◽

Binding Groove ◽

The Many ◽

The Cost

The COOH-terminal peptides of pigeon and moth cytochrome c, bound to mouse IEk, are two of the most thoroughly studied T cell antigens. We have solved the crystal structures of the moth peptide and a weak agonist–antagonist variant of the pigeon peptide bound to IEk. The moth peptide and all other peptides whose structures have been solved bound to IEk, have a lysine filling the p9 pocket of IEk. However, the pigeon peptide has an alanine at p9 shifting the lysine to p10. Rather than kinking to place the lysine in the anchor pocket, the pigeon peptide takes the extended course through the binding groove, which is characteristic of all other peptides bound to major histocompatibility complex (MHC) class II. Thus, unlike MHC class I, in which peptides often kink to place optimally anchoring side chains, MHC class II imposes an extended peptide conformation even at the cost of a highly conserved anchor residue. The substitution of Ser for Thr at p8 in the variant pigeon peptide induces no detectable surface change other than the loss of the side chain methyl group, despite the dramatic change in recognition by T cells. Finally, these structures can be used to interpret the many published mutational studies of these ligands and the T cell receptors that recognize them.

Download Full-text

Design of more potent quinazoline derivatives as EGFRWT inhibitors for the treatment of NSCLC: a computational approach

Future Journal of Pharmaceutical Sciences ◽

10.1186/s43094-021-00279-3 ◽

2021 ◽

Vol 7 (1) ◽

Author(s):

Muhammad Tukur Ibrahim ◽

Adamu Uzairu ◽

Sani Uba ◽

Gideon Adamu Shallangwa

Keyword(s):

Lung Cancer ◽

Molecular Docking ◽

Binding Affinity ◽

Positive Control ◽

Computational Approach ◽

Tyrosine Kinase Receptor ◽

Binding Affinities ◽

Binding Interaction ◽

Pharmacokinetic Properties ◽

Inhibitory Activities

Abstract Background Lung cancer remains the leading and deadly type of cancer worldwide. It was estimated to account for about 25% of the 7 million people that died as a result of cancer-related issues/mortality every year in the world. Non-small cell lung cancer (NSCLC) is the lethal/deadly class of lung cancer with nearly 1.5 million reported cases and less than 20% survival rate. Therefore, it becomes necessary to explore more effective NSCLC drugs. Result A computational approach was employed here to design ten new EGFRWT inhibitors using compound 18 as a template for the design identified with the best binding affinity and good pharmacokinetic properties previously reported in our work. The modeled inhibitory activities of these newly designed EGFRWT inhibitors (range from 7.746966 to 11.09261) were better than that of the hit compound with pIC50 of 7.5639 and gefitinib the positive control with pIC50 of 5.879426. The ligand-binding interaction between these newly designed EGFRWT inhibitors and the EGFR tyrosine kinase receptor as shown in Table 3 was investigated and elucidated using molecular docking protocol. Based on the molecular docking results, the binding affinities of these newly designed EGFRWT inhibitors were found to be between − 8.8 and − 9.5 kcal/mol. The designed compound SFD10 has the highest binding affinity of − 9.5 kcal/mol followed by compound SFD8 (with a binding affinity of − 9.3 kcal/mol), then by compound SFD9 and 4 (each with a binding affinity of − 9.3 kcal/mol). None of them was found to have more than one violation of the filtering criterion used in this study thereby showing good ADMET properties. Conclusion The modeled inhibitory activities and binding affinities of these newly designed EGFRWT inhibitors were found to be higher than that of the template compound and the control (gefitinib) used in this research. They were also seen to be non-toxic with good pharmacokinetic properties.

Download Full-text

BiodMHC: an online server for the prediction of MHC class II-peptide binding affinity

Journal of Genetics and Genomics ◽

10.1016/s1673-8527(08)60117-4 ◽

2009 ◽

Vol 36 (5) ◽

pp. 289-296 ◽

Cited By ~ 2

Author(s):

Lian Wang ◽

Danling Pan ◽

Xihao Hu ◽

Jinyu Xiao ◽

Yangyang Gao ◽

...

Keyword(s):

Binding Affinity ◽

Mhc Class Ii ◽

Peptide Binding ◽

Class Ii ◽

Peptide Binding Affinity

Download Full-text

Toward Guided Mutagenesis: Gaussian Process Regression Predicts MHC Class II Antigen Mutant Binding

Journal of Chemical Information and Modeling ◽

10.1021/acs.jcim.1c00458 ◽

2021 ◽

Author(s):

David R. Bell ◽

Serena H. Chen

Keyword(s):

Gaussian Process ◽

Mhc Class Ii ◽

Gaussian Process Regression ◽

Class Ii ◽

Class Ii Antigen ◽

Mhc Class Ii Antigen

Download Full-text

SARS-CoV-2 Virus and Human Leukocyte Antigen (HLA) Class II: Investigation in silico of Binding Affinities for COVID-19 Protection and Vaccine Development

Journal of Immunological Sciences ◽

10.29245/2578-3009/2020/4.1198 ◽

2020 ◽

Vol 4 (4) ◽

pp. 12-23 ◽

Cited By ~ 2

Author(s):

Spyros Charonis ◽

Effie-Photini Tsilibary ◽

Apostolos Georgopoulos

Keyword(s):

Human Leukocyte Antigen ◽

Binding Affinity ◽

In Silico ◽

Vaccine Development ◽

Human Leukocyte ◽

Class Ii ◽

Hla Class Ii ◽

Effective Vaccine ◽

Binding Affinities ◽

Leukocyte Antigen

SARS-CoV-2 causes COVID-19, urgently requiring the development of effective vaccine(s). Much of current efforts focus on the SARS-CoV-2 spike-glycoprotein by identifying highly antigenic epitopes as good vaccine candidates. However, high antigenicity is not sufficient, since the activation of relevant T cells depends on the presence of the complex of the antigen with a suitably matching Human Leukocyte Antigen (HLA) Class II molecule, not the antigen alone: in the absence of such a match, even a highly antigenic epitope in vitro will not elicit antibody formation in vivo. Here we assessed systematically in silico the binding affinity of epitopes of the spike-glycoprotein to 66 common HLA-Class-II alleles (frequency ≥ 0.01). We used a sliding epitope window of 22-amino-acid-width to scan the entire protein and determined the binding affinity of each subsequence to each HLA allele. DPB1 had highest binding affinities, followed by DRB1 and DQB1. Higher binding affinities were concentrated in the initial part of the glycoprotein (S1-S460), with a peak at S223-S238. This region would be well suited for effective vaccine development by ensuring high probability for successful matching of the vaccine antigen from that region to a HLA Class II molecule for CD4+ T cell activation by the antigen-HLA molecule complex.

Download Full-text