Using multi-omics to investigate the host response to COVID-19

Abstract

Many individuals with coronavirus disease 2019 (COVID-19) are asymptomatic or have only mild symptoms. However, a subset of patients experience severe disease, which can progress to respiratory distress and organ failure. End-stage kidney disease (ESKD) patients are at a particularly high risk of fatal COVID-19. Despite this, most ESKD patients could not fully self-isolate during the UK national lockdowns as they required regular haemodialysis treatments. As a result, there was a unique opportunity for longitudinal sampling of both outpatients and inpatients with COVID-19 in a group at high risk of severe disease. New technologies, that permit the concurrent measurement of thousands of molecular traits, have the potential to grant us an improved understanding of the mechanisms underlying severe COVID-19 and could help prioritise new therapeutic targets. Blood transcriptomics and plasma proteomic profiling were applied to two cohorts of ESKD patients with COVID-19. Using these data, I characterised the temporal profiles of molecules following COVID-19 onset. For instance, LRRC15, a proposed receptor for SARS-CoV-2, fell in abundance in patients with severe disease but remained relatively stable in milder cases. Samples were also collected from patients following clinical recovery from COVID-19, which revealed that pathways related to clotting and inflammation remain dysregulated for months after infection by SARS-CoV-2. These cohorts were collected as part of an observational study, so I could not discern between molecules that drive disease, versus those that are simply downstream effects of clinical deterioration. As a follow-up, Mendelian randomisation was performed. This involved the use of genetic variants as instruments to identify proteins that may cause severe COVID-19. While this approach did not conclusively identify proteins that mediate COVID-19 outcomes, it did reveal a mechanism by which differential splicing of the transmembrane region of the Fas cell death receptor influenced its abundance in plasma.