Marion Humbert[1], Anna Olofsson[1], David Wullimann[2], Julia Niessl[2], Emma Holdcroft[3], […], Linda Björkem-Bergman[4], Maria C. Jenmalm[5], Hans-Gustaf Ljunggreen[2], Marcus Buggert[2], Annika C. Karlsson[1]

Affiliates

Division of Clinical Microbiology, Department of Laboratory Medicine, Karolinska Institutet, Stockholm, Sweden[1]; Center for Infectious Medicine, Department of Medicine Huddinge, Karolinska Institutet, Stockholm, Sweden[2]; Biozentrum, University of Basel, Switzerland[3]; Division of Clinical Geriatrics, Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden[4]; Division of Inflammation and Infection, Department of Biomedical and Clinical Sciences, Linköping University, Sweden[5].

Abstract

Background. Human coronaviruses (HCoVs) have been suggested by several studies to promote cross-reactive T cells recognizing SARS-CoV-2 in unexposed individuals, which role in disease control remains debated. Our study aimed at comparing the cellular immunity to HCoV-OC43 with the cross-reactive CD4+ T cell response against SARS-CoV-2 (Aim I), and at characterizing OC43-specific T cell responses throughout life (Aim II).

Methods. CD4+ T cell responses against OC43 and SARS-CoV-2 peptide megapools (MPs) were evaluated in a cohort of blood donors (Aim I), as well as in children at 6 years old (y.o.), younger adults (26-58 y.o.) and older adults (60-83 y.o.) (Aim II). All donors were SARS-CoV-2 seronegative. Antigen-specific memory (m)CD4+ T cells were characterized after a 10h-stimulation with MPs – Spike (S) or Nucleocapsid (N) -, thanks to a flow cytometry panel combining activation-induced marker (AIM) and intracellular cytokine staining (ICS) assays. Principal component analysis (PCA) and uniform manifold approximation and projection (UMAP) for dimension reduction were performed.

Results. Aim I. Cross-reactive SARS-CoV-2-specific mCD4+ T cell responses were mainly directed to the S region, with significantly lower magnitudes towards N. The magnitude of the cross-reactive SARS-CoV-2 S-specific mCD4+ T cell response was significantly lower, compared with OC43 S. A close correlation was identified between the cross-reactive SARS-CoV-2 S-, and N-, compared to OC43 S-, and N-specific T cell responses. There was no significant difference in the overall functional or phenotypic profiles of the OC43 S- and cross-reactive SARS-CoV-2 S-specific mCD4+ T cell populations. However, we observed a significant decrease in the magnitude of IFN-g production by cross-reactive SARS-CoV-2 S-specific mCD4+ T cell, compared with OC43 S. Several combinations of mono- and poly-functional T cell subsets separated OC43 N- from cross-reactive SARS-CoV-2 N-specific mCD4+ T cells, showing their highly limited functional ability.
Aim II. When comparing younger adults and older adults, a detailed functional and phenotypic analysis revealed substantial differences in the mCD4+ T cell responses against OC43 S, OC43 N, and SARS-CoV-2 S. A clear segregation between the two age groups was demonstrated using PCA, and was driven by significantly decreased levels of IFN-g, IL-2 and TNF-a in older adults.
Using our three age cohorts (6-83 y.o.), we observed a decrease in the frequencies of detectable mCD4+ T cells with the ability to respond to OC43 S, OC43 N, and SARS-CoV-2 S with increasing age. Using UMAP and Phenograph, we observed a cytokine-enriched cluster (TNF-a+ IL-2+ IFN-g+) that contained most antigen-specific cells, this fraction being the lowest amongst the older adult group. The distribution of cells was also strongly influenced by the expression of CD38, with a tendency for the 6 y.o. to have more CD38+ cells, compared with the older adult group ( > 60 y.o.). Importantly, the cluster distribution of OC43 S- and SARS-CoV-2 S-specific mCD4+ T cells from each age group was similar.

Conclusions. The functional and phenotypic qualities of cross-reactive SARS-CoV-2 S-specific mCD4+ T cell responses mimic T cell immunity against OC43 S. In addition, diminished T cell immunity against OC43 and SARS-CoV-2 was associated with older age. We also observed a peculiar phenotype in 6 y.o, compared with other age groups. These findings may partly explain why the severity of COVID-19 increases with age.