A multimodal atlas of COVID-19 severity identifies hallmarks of dysregulated immunity 2258066

Abstract

Abstract Introduction The alpha-variant wave of the COVID-19 pandemic provided a unique opportunity to study, at single-cell resolution, how near-universal exposure to the same pathogen can lead to either effective or dysfunctional immune responses in humans. Methods We analyzed 2.5 million circulating immune cells from 428 patients across time points (840 PBMC samples), encompassing three contemporaneous SARS-CoV-2 cohorts: acutely infected patients at five WHO disease severity levels and three time points, patients from the first randomized control trial to study efficacy of tocilizumab in management of COVID-19, and convalescent patients three months after infection. We used linear modeling to integrate multiple data types – single-cell RNA-seq, CITE-seq, TCR and BCR sequencing, viral load measurements, viral neutralization assays, detection of 75 autoantibodies, HLA genotype data, and serum proteomics covering 1,463 targets – to derive the most comprehensive view to-date of the biological features of COVID-19 disease severity. Results We show that myeloid-derived suppressor cells (MDSCs) act as a key immunologic pivot point in severe COVID-19. Myeloid dysfunction is marked by impaired antigen presentation and drives a non-productive adaptive immune response. Severe disease is also linked to autoantibodies targeting type I interferons, specific HLA-DQB1 allelic variants, and serum IL-6 levels. Tocilizumab treatment eliminates CLU-expressing MDSCs and ISG-positive myeloid subsets, restores antigen presentation, and reactivates productive adaptive immunity. In convalescence 3-months post-infection, we found persistently high ICOS expression in regulatory T cells. Conclusion Overall, we define distinct innate and adaptive host immune responses associated with acute, IL-6—responsive, and convalescent SARS-CoV-2 infection. Our multimodal and high-dimensional dataset with curated clinical metadata provides a foundational and clinically relevant resource for modeling host immune response biology in health and disease. Funding Source We acknowledge the following funding sources: this work was supported by several training grants, including a NIAID grant T32AR007258 (to KS), three NHLBI grants 5T32HL116275-13 (to CC), 5T32HL129970-09 (to APN), and the K08HL157725 (to PS), as well as the American Heart Association Career Development Award (to PS). PS was also supported by the Brigham and Women’s Hospital Innovation Evergreen Fund. EY was supported by funding from the Stanford Medical Scholars program. RJX acknowledges supports from NIH DK43351 and U19AI142784. RJX and AR were supported by the Manton Foundation and the Klarman Cell Observatory. PJU was supported by Third Rock Ventures; Henry Gustav Floren Trust; Stanford Department of Medicine Team Science Program; Stanford Medicine Office of the Dean; and National Institutes of Health R01 grants AI175771 and AI182319-02. RPB acknowledges funding support from the Massachusetts General Hospital Executive Committee on Research, the American Lung Association, and the Broad Institute’s Next Generation Scholar award. MBG, MRF, and NH were supported by an American Lung Association COVID-19 Action Initiative grant. MBG and MRF were supported by a grant from the Executive Committee on Research at MGH. NH acknowledges was supported by NIH/NIAID U19 AI082630, a Chair and gift from Sandra, Sarah, and Arthur Irving. ACV acknowledges funding support from the COVID-19 Clinical Trials Pilot grant from the Executive Committee on Research at MGH; a COVID-19 Chan Zuckerberg Initiative grant (2020-216954); the funds from the Manton Foundation and the Klarman Family Foundation; the Broad Institute’s Next Generation Scholar award; the MGH Howard M. Goodman Fellowship; the National Institutes of Health (DP2CA247831); work at the Broad Institute was supported by a gift from an anonymous donor. Topic Categories Computational and Systems Immunology (COMP)