Experimental validation of immunogenic SARS-CoV-2 T cell epitopes identified by artificial intelligence

Abstract

During the COVID-19 pandemic we utilized an AI-driven T cell epitope prediction tool, the NEC Immune Profiler (NIP) to scrutinize and predict regions of T cell immunogenicity (hotspots) from the entire SARS-CoV-2 viral proteome. These immunogenic regions offer potential for the development of universally protective T cell vaccine candidates. Here, we validated and characterized T cell responses to a set of minimal epitopes from these AI-identified universal hotspots. Utilizing a flow cytometry-based T cell activation-induced marker (AIM) assay, we identified 59 validated screening hits, of which 56% (33 peptides) have not been previously reported. Notably, we found that most of these novel epitopes were derived from the non-spike regions of SARS-CoV-2 (Orf1ab, Orf3a, and E). In addition, ex vivo stimulation with NIP-predicted peptides from the spike protein elicited CD8+ T cell response in PBMC isolated from most vaccinated donors. Our data confirm the predictive accuracy of AI platforms modelling bona fide immunogenicity and provide a novel framework for the evaluation of vaccine-induced T cell responses.

People who use drugs show no increase in pre-existing T-cell cross-reactivity toward SARS-CoV-2 but develop a normal polyfunctional T-cell response after standard mRNA vaccination

“…To investigate the response of CD8 + cytotoxic T lymphocytes (CTL), we stimulated the cells with the Peptivator mix or a pool of 23 9/10-mer spike peptides (the NOI pool) that were predicted by the NEC Immune profiler AI tool (see Supplementary Methods for peptide list) ( 43 ) and experimentally validated ( 41 ). We determined the frequency of spike-specific T cells after stimulation with either the Peptivator mix or the NOI pool by measuring CD8 CTL cell responses using the Boolean “Or” gating on the same double positive regions defined for the analysis of the Th cell subset.…”

“…To test SARS-CoV-2-specific reactivity of T cells from pre-vaccinated donors, PBMC from each patient were distributed on a 96-well round bottom cell culture plate at a concentration of 3×10 5 cells per well (~5×10 4 and ~1×10 5 CD8 + and CD4 + T cells, respectively) and stimulated with a pool of SARS-CoV-2 peptides (one peptide per well at 1.5 µg/mL: final volume = 200 µL per well; for the list of the 9–10-mer peptides used in these studies, see Supplementary Methods ) ( 41 , 42 ). At day 3, supernatants (SN) were collected for cytokine analysis (ELISA), after which methyl- 3 H thymidine (Hartmann Analytic GmbH, 0.02 mCi/mL) was added for cell proliferation assessment.…”

“…Among these peptides, 43 derived from the spike protein and 58 from non-spike regions (see Supplementary Methods ). Additionally, a pool of 23 9/10-mer spike peptides (the NOI pool) validated in our laboratory ( 41 , 42 ) was used to identify vaccine response. The list of peptides is shown in Supplementary Methods .…”

An Exploratory Bioinformatic Investigation of Cats’ Susceptibility to Coronavirus-Deriving Epitopes

“…FIPV-deriving epitopes were instead retrieved in canine enteric coronavirus (CCoV) and porcine transmissible gastroenteritis coronavirus (TGEV), confirming the widely demonstrated close relationship between these three coronaviruses [ 45 ]. To validate the six predicted sequences with experimental data, we performed a search for the obtained epitopes on the immune epitope database IEDB ( , accessed on 22 February 2024) [ 46 ], which reported that (i) sars-r Ep1 (IEDB ID: 2249103) has been identified by an AI prediction in a pool of sequences as an epitope for several class I HLA alleles and tested in a mix of peptides that acted as a CD8 + T-cell activator [ 47 ], (ii) sars-s Ep5 (IEDB ID: 1332221) has been found to bind in vitro different alleles of the HLA-I B locus [ 41 ], and (iii) sars-s Ep6 (IEDB ID: 1323461) has been extensively studied among HLA-I-binding restricted peptides for its ability to activate CD8 + T cells [ 48 , 49 , 50 , 51 , 52 , 53 ]. On the other hand, FIPV-derived fipv-r Ep4 and fipv-s Ep8 epitopes have not been found in the database, while the fipv-s Ep9 sequence is contained in a longer epitope (IEDB ID: 142156), which has been found to significantly enhance the feline interferon (IFN)-γ levels in peripheral blood mononuclear cells (PBMC) and has been identified as an antibody-binding epitope [ 54 ].…”

An Exploratory Bio-Informatic Investigation of Cats’ Susceptibility to Coronavirus-Deriving Epitopes

“…FIPVderiving epitopes were instead retrieved in canine enteric coronavirus (CCoV) and porcine transmissible gastroenteritis coronavirus (TGEV), confirming the widely demonstrated close relationship between these three coronaviruses [45]. To validate the six predicted sequences with experimental data, we performed a search for the obtained epitopes on the immune epitope database IEDB (www.iedb.org, accessed on 22 February 2024) [46], which reported that (i) sars-r Ep1 (IEDB ID: 2249103) has been identified by an AI prediction in a pool of sequences as an epitope for several class I HLA alleles and tested in a mix of peptides that acted as a CD8 + T-cell activator [47], (ii) sars-s Ep5 (IEDB ID: 1332221) has been found to bind in vitro different alleles of the HLA-I B locus [41], and (iii) sars-s Ep6 (IEDB ID: 1323461) has been extensively studied among HLA-I-binding restricted peptides for its ability to activate CD8 + T cells [48][49][50][51][52][53]. On the other hand, FIPV-derived fipv-r Ep4 and fipv-s Ep8 epitopes have not been found in the database, while the fipv-s Ep9 sequence is contained in a longer epitope (IEDB ID: 142156), which has been found to significantly enhance the feline interferon (IFN)-γ levels in peripheral blood mononuclear cells (PBMC) and has been identified as an antibody-binding epitope [54].…”