Patterns of restricted TCR usage following SARS-CoV-2 vaccination and severe disease

Emily Parsons¹Zhongyan Lu^1,2Stephanie A Richards^2,3Amanda Zelkoski^1,2Janifer Le^1,2Naraen Palanikumar^1,2Phuong Nguyen^1,2Camille Alba^2,4Gauthaman Sukumar^2,4John Rosenberger^2,4Xijun Zhang^2,4Timothy H Burgess³Rhonda Colombo^2,3,5,6Katrin Mende^2,3,7Catherine Berjohn^2,5,8Nursat Epsi^2,3Brian K Agan^2,3David Tribble³David A Lindholm^5,7Clifton L Dalgard⁴Simon D Pollett^2,3Allison Marie Wahl Malloy^1*

• ¹Department of Pediatrics, Uniformed Services University of the Health Sciences, Bethesda, United States
• ²Henry M Jackson Foundation for the Advancement of Military Medicine (HJF), Bethesda, Maryland, United States
• ³Infectious Disease Clinical Research Program, Department of Preventive Medicine and Biostatistics, Uniformed Services University of the Health Sciences, Bethesda, MD, United States
• ⁴Department of Anatomy, Physiology and Genetics, Uniformed Services University, Bethesda, Maryland, United States
• ⁵Department of Medicine, Uniformed Services University of the Health Sciences, Bethesda, Illinois, United States
• ⁶Madigan Army Medical Center, Tacoma, Washington, United States
• ⁷Brooke Army Medical Center, San Antonio, Texas, United States
• ⁸Naval Medical Center San Diego, San Diego, California, United States

T cells influence COVID-19 severity and establish long-lasting immune memory in response to vaccination and infection. The diversity of the T cell repertoire, and complexity of T cell epitope recognition, make it challenging to define protective epitope-specific T cells. In this study, we created a highly specific TCR meta-database to identify T cell epitopes from the nearly complete SARS-CoV-2 proteome and determine whether vaccination with mRNA vaccines influenced the TCR repertoire. Using this database, we analyzed immunosequencing data of genomic DNA to define the variable region of T cell receptor (TCR) β chain (TCRB) sequences among participants in a longitudinal COVID-19 cohort study. The TCR repertoire was compared between participants who were vaccinated or unvaccinated against SARS-CoV-2 and stratified by disease severity. TCR diversity was measured using clonality, an index defined as the inverted normalized Shannon entropy. Our analysis demonstrated that highly clonal TCR repertoires correlated with age and comorbidities. Using our meta-database approach, we found that vaccinated participants hospitalized with infection had the most restricted SARS-CoV-2specific CD8 TCR repertoire. However, TCRB with predicted specificity to non-spike SARS-CoV-2 proteins dominated the response, even in vaccinated participants. We identified a peptide sequence in the ORF10 accessory protein that was more frequently recognized in study participants with mild disease. Conversely, CD8 T cell recognition of a peptide sequence in ORF1ab more closely correlated with severe disease. Overarchingly, TCR repertoire analysis revealed that CD8 T cells responding to SARS-CoV-2 broadly recognize epitopes across the SARS-CoV-2 proteome, and provided opportunities to identify epitopes associated with disease.