Simon Mallal

Introduction: Isolevuglandin (IsoLG)-adducted peptides contribute to hypertension by enhancing class I human leukocyte antigen (HLA)-mediated T cell activation. The capacity of different HLA alleles to present IsoLG-modified peptides varies, influencing immune activation and disease progression. To elucidate the molecular determinants of peptide-HLA (pHLA) interactions, we combined cell-based peptide binding assays, structural modeling, and deep learning to predict pHLA binding affinity (IC50) and the presentation of IsoLG-modified epitopes.

Methods: IsoLG presentation was quantified in K562 cells, each transfected to express a single HLA (45 total, covering 90% of the US population), using flow cytometry-based Förster resonance energy transfer (FRET), which measures peptide-HLA proximity via donor-acceptor fluorescence. In computational studies, we filtered experimental pHLA binding data from the Immune Epitope Databank (IEDB) and modeled 49,268 structures using Rosetta and AlphaFold. Graph embeddings were constructed for pHLA complexes and unbound HLA alleles, incorporating per-residue and pairwise binding energies, and IC50 values. HLA embeddings were enriched with IsoLG FRET data to train a graph neural network, which combines information from both local molecular structure and broader interaction patterns using graph convolution and self-attention. The approach accommodates missing data and variable graph sizes using custom loss (error tracking) functions to manage outliers and incomplete measurements.

Results: Using FRET assays, HLA alleles were grouped by isoLG adduct presentation: high (red), medium (orange), and low (blue). Notably, baseline FRET correlated with tBHP-stimulated IsoLG-adduct levels (Pearson r = 0.907), suggesting the presence of pre-existing IsoLG adducts (Figure 1A). After 100 training epochs (cycles), the model's prediction error (loss) stabilized below 0.026 (Figure 1B). On the test set, the model's predictions were typically within 13-16% of the normalized log(IC50) value (RMSE = 0.155; MAE = 0.134).

Conclusions: This study establishes a framework combining empirical binding assays, structural modeling, and graph-based deep learning for immunogenicity prediction in hypertension. The architecture shows promise in identifying relevant interaction sites between peptides and the HLA binding groove. Potent IsoLG-adduct presenting HLAs may represent a high-risk group for immune-mediated hypertension and related diseases.

Introduction: Isolevuglandins (isoLGs) are lipid peroxidation products that covalently modify lysine residues on self-proteins. These modified self-proteins are processed to peptides that are presented in the context of class I major histocompatibility complexes (MHC-I). In hypertension, this modification enhances CD8+ T cell activation, contributing to inflammation and elevated blood pressure. In mice we found that IsoLG-adduct presentation is highly dependent on class I MHC structure. We sought to determine if human class I MHC (HLAs) exhibit similar variability in their ability to present these adducts by screening alleles representing 90% of the population.

Hypothesis: We hypothesize that HLA subtypes exhibit variable capacity for isoLG presentation.

Methods: HLA alleles representing 90% demographic frequency were identified from the US National Merit Donor Program Allele Frequency Database. HLA-null human K562 cells were transduced with these alleles, expanded, and enriched for HLA expression by flow cytometry to produce single HLA allele lines. To induce isoLG adduct presentation, cells were treated with 1mM tert-Butyl hydroperoxide (tBHP). After 24 hours, we employed Forster resonance energy transfer (FRET) between HLAs and isoLG-adducts using antibodies specific for both, which we have shown is indicative of IsoLG-adduct presentation (Fig 1A-C).

Results: Substantial variability for isoLG-adduct presentation was observed between HLA subtypes A, B and C (Figure 1A-1C). Scikit-learn’s KMeans clustering revealed four patterns of isoLG-adduct presentation among HLAs: Low baseline/low activation, low baseline/high activation, intermediate baseline/intermediate activation, and high baseline/high activation alleles (Fig 1D).

Conclusions: These findings identify specific HLA alleles as high isoLG presenters, potentially associated with increased immune activation risk. Combining these findings, HLA alleles, and diagnosis code databases could aid in identifying high-risk haplotypes for immune activation in human hypertension.