Biology subjects

Soria-Bustos, J.

Publications and source records attributed to Soria-Bustos, J..

4 recordsLinked to original sources

bioRxiv2025

Learned Geometry, Predicted Binding: Structurally-BasedPrediction of Peptide:MHC Binding Using AlphaFold 3 EnablesCD4 T Cell Epitope Prediction

The accurate prediction of T cell epitope peptides within proteins of interest has a wide range of applications, but is complicated by the multiple determinants of antigenicity, the polymorphism of the Major Histocompatibility Complex (MHC) locus, and the great diversity of possible peptide antigens. Leading in silico methods use a variety of statistical approaches to learn from sequences identified through both in vitro MHC binding and peptide elution studies, but their performance remains imperfect, particularly for MHC II-restricted responses. Here we present MHCIIFold-GNN, an entirely orthogonal solution to this problem that combines three new elements: (i) a highly-multiplexed peptide:MHCII binding assay, (ii) generalizable structural modeling using AlphaFold3, and (iii) transfer learning with a Graph-based Neural Network. Trained exclusively on newly-generated in vitro binders, we show that MHCIIFold-GNN enables state-of-the-art prediction of CD4 T cell epitopes presented by diverse MHC II proteins, on par with non-structural methods that rely on much larger datasets including naturally-processed ligands. Moreover, when MHCIIFold-GNN and a leading non-structural method are combined, we observe unparalleled performance on a held-out test set (11 % boost), underscoring the orthogonality of the methods. These results highlight the power of a new class of structure-informed approaches to the CD4 T cell epitope prediction problem.

bioinformatics↗

bioRxiv2024

Preclinical Proof of Concept for SNAPTM-TIL: a personalized, next-generation tumor-infiltrating lymphocyte therapy platform

Tumor-infiltrating lymphocyte (TIL) therapy, which involves extracting, expanding, and reinfusing immune cells to target cancer cells, has shown promise in melanoma treatment, but requires optimization for broader efficacy. The success of TIL therapy depends on the recognition of tumor-associated antigens, but neoantigen-reactive T-cells are often rare and exhausted in less immunogenic malignancies. Isolating T cells enriched in neoantigen reactivity prior to in vitro expansion and reinfusion may improve the response rates. To this end, our proprietary Specific Neo-Antigen Peptides (SNAP) technology platform improves the accuracy of neoantigen prediction and validation by combining advanced computational modeling and PepSeq, a high-throughput screen for the physical credentialing of putative neoantigens based on their affinity to bind patient-specific HLA class II proteins. This approach allows for the education and enrichment of TILs (SNAP-TILs) with personalized, predefined, highly immunogenic neoantigens prior to expansion. Using the SNAP platform, we consistently achieved, on average, an SNAP-TIL product comprising 96% CD3+ cells, with a mixture of 75% effector and 23% central memory cells. SNAP-TILs exhibited greater efficacy and selectivity in immune infiltration than TIL, which was expanded by the rapid expansion protocol alone using ex vivo models. SNAP-TIL was also reactive in highly and poorly immunogenic tumors, with 70% and 50% tumor growth inhibition in melanoma and pancreatic patient-derived xenograft models, respectively. This study demonstrates the novel benefit of our Personalized Neoantigen Pipeline approach, potentially providing a durable antitumor immune response for a larger proportion of cancer patients.

immunology↗

bioRxiv2024

Integrated, high-dimensional analysis of CD4 T cell epitope specificities and phenotypes reveals unexpected diversity in the response to Mycobacterium tuberculosis

Immunity to Mycobacterium tuberculosis (Mtb), like many pathogens, is encoded jointly by the antigen specificities and functions of responding CD4 T cells. However, these features span a large two-dimensional possibility space - defined on one axis by the Mtb proteome, and on the other by the T cell transcriptome - that exceeds the dimensionality of existing technologies. Here we present an approach ("CRESTA") that combines highly-multiplexed DNA-barcoded epitope probes, single cell sequencing, and clonal analysis of T Cell Receptors (TCRs) to robustly detect rare antigen-specific CD4 T cells across hundreds of epitopes simultaneously and reveal their transcriptome-wide phenotypes. By comprehensively assaying known epitopes in Mtb-infected participants, we reveal polyclonal and multi-epitope responses across a spectrum of differentiation states, uncover previously-unobserved phenotypic diversity within and between epitopes, and increase the total number of known Mtb epitope-mapped TCR:{beta}s by [~]8-fold. We expect CRESTA to enable high-dimensional analyses of CD4 T cell responses in various settings, including infection, cancer, autoimmunity and allergy.

immunology↗

bioRxiv2024

The transcriptional regulator Lrp activates the expression of genes involved in tilivalline enterotoxin biosynthesis in Klebsiella oxytoca

The toxigenic Klebsiella oxytoca strains secret the tilivalline enterotoxin, which causes antibiotic-associated hemorrhagic colitis. The tilivalline is a non-ribosomal peptide synthesized by enzymes encoded in two divergent operons clustered in a pathogenicity island. The transcriptional regulator Lrp (leucine-responsive regulatory protein) controls the expression of several bacterial genes involved in virulence. In this work, we determined the transcriptional expression of aroX and npsA, the first genes of each tilivalline biosynthetic operon in K. oxytoca MIT 09-7231 wild-type and its derivatives {Delta}lrp mutant and complemented strains. The results show that Lrp directly activates the transcription of both aroX and npsA genes by binding to the intergenic regulatory region in a leucine-dependent manner. Furthermore, the lack of Lrp significantly diminished the cytotoxicity of K. oxytoca on HeLa cells due to tilivalline reduced production. Altogether, our data highlight Lrp as a new regulator by which cytotoxin-producing K. oxytoca strains control the expression of genes involved in the biosynthesis of their main virulence factor. IMPORTANCETilivalline is an enterotoxin that is a hallmark for the cytotoxin-producing K. oxytoca strains, which cause antibiotic-associated hemorrhagic colitis. The biosynthesis of tilivalline is driven by enzymes encoded by the aroX- and NRPS-operons. In this study, we discovered that the transcriptional regulator Lrp directly activates expression of the aroX- and NRPS-operons and, in turn, tilivalline biosynthesis. Our results underscore a molecular mechanism by which tilivalline production by toxigenic K. oxytoca strains is regulated and shed further light on developing strategies to prevent the intestinal illness caused by this enteric pathogen.

molecular biology↗