Defining high-resolution T-cell correlates of protection in abortive versus seropositive SARS-CoV-2 infection

We have recently demonstrated that exposure to SARS-CoV-2 without detectable infection (repeatedly PCR- antibody-) expands T-cell responses in seronegative health care workers (HCW) who have an innate signature compatible with a subclinical/abortive infection (Swadling.Nature.2022). Importantly, we show in this unique setting of HCW who resist overt infection, that their T-cells preferentially target the non-structural regions of SARS-CoV-2, in particular essential proteins within the replication-transcription complex (RTC; NSP12 polymerase, NSP7 cofactor, NSP13 helicase). In contrast, T-cell responses in HCW who have overt seropositive infection predominantly target structural regions of the virus.

Having newly identified an association between expansion of RTC-specific T-cells and early viral control, we will now characterise these T-cells at greater resolution to determine if there are specific qualities, other than specificity, which could explain their ability to mediate early viral control.

We will investigate both pre-existing cross-reactive T-cells, recruited into the immune response, and T-cells generated de novo during abortive infection, and contrast these with T-cells responses during overt SARS-CoV-2 infection.

Part-I: This will include functional assessment of SARS-CoV-2-reactive T-cells via spectral cytometry, which allows multiparameter single cell analysis, maximising insights from rare samples. MHC class I/II multimers and Activation-induced marker assay (AIM) will be used to investigate T-cell phenotypes, including: memory subset, trafficking, transcription factor profile, coinhibitory and costimulatory receptors. High-dimensional data will require dimension reduction and emerging data visualisation tools, such as UMAP/tSNE/SPADE/Flowsom.

Part-II: In addition to comprehensive assessment of the quality of the T-cell response during abortive infection, we will further refine the specificity of the T-cells correlating with protection from detectable SARS-COV-2 infection. Epitopes within the RTC region will be identified using cutting-edge machine learning tools, such as those develop by collaborator Dr Bravi (Imperial Mathematics), RBM-MHC. These epitopes will be validated and their association with abortive infection determined through cohort screening (COVIDSortium, n=731 ~70 abortive infections ~140 lab-confirmed infections).

Part-III: Finally, immunodominant epitopes will be identified using TCR-repertoire analysis, through the creation of an RTC-T-cell receptor library, generated by expanding T-cells using RTC peptides and sequencing TCRs. Importantly, TCRs will be clustered into metaclonotypes according to sequenced similarity, determined by machine-learning approach with Dr Tiffeau-Mayer (UCL, IIT), allowing inter-individual comparisons of the TCRs targeting the RTC. Publicly available TCR repertoires will also be mined to see if which RTC-specific TCRs are expanded in pre-pandemic repertoires, showing they are cross-reactive with coronaviruses circulating before the pandemic, and repertoires from bronchoalveolar lavage samples, showing enrichment at the site of coronavirus control. This work will ultimately identify if specific TCR sequence motifs or TCR-epitope pairs correlate with protection from detectable SARS-CoV-2 infection.

Overall, we will take a holistic approach, examining the fine epitope and TCR specificity and functional and phenotypic characteristics of T-cells targeting the RTC of SARS-CoV-2 to refine this correlate of protection, leading to a greater understanding of the role of T cells in viral control and
informing the type of T-cell immunity that should be aimed for with next generation vaccines.