What happens to CD8+ memory cells a year after infection?

Since the beginning of the COVID-19 pandemic, biomedical research has tried to better understand long-term immunity against SARS-CoV-2. At the beginning of infection, T-CD8 cells proliferate, and activate as effector cells upon contact with viral antigens, to eventually eliminate infected cells from the body. In the case of COVID-19, we do not understand their evolution, or which populations become memory cells that will protect the host in future reinfections.  

In order to better understand these cellular dynamics, Swiss researchers at Zurich University enrolled 175 COVID-19 patients from whom they took blood samples over a year to analyse T-CD8 cells using several approaches. They concentrated on patients carrying certain major histocompatibility complex alleles (MHC-I : HLA-A*01:01, HLA-A*11:01).

In the majority of patients, T-CD8 cells specific to SARS-CoV-2 were detected up to a year after the acute phase of the infection, with strong heterogeneity in the effector (T_EFF) and memory cells. By examining the genetic profiles of these latter cells, the researchers identified 12 different sub-types of T-CD8 cells (classified from 1 to 12).

During the acute phase of infection, the activated cells were in the majority Ki-67+ and HLA-DR+. There was an enrichment of cells expressing CD62L, the granzyme B (which perforated infected cells) and the CX3CR1 marker (in particular in severe cases). The sub-types 1,2 and 12 dominated during this phase and corresponded to proliferating and cytotoxic activated T-CD8 cells. The researchers defined a genetic signature specific to the acute phase (the genes NKG7, PRF1, GZMB, CENPU/F, MK167), which diminishes over time. These T_EFF cells are specific to numerous viral antigens and show strong cytokines/TCR signalling.

Between 6 months and one year after the acute phase, the majority of these cells declined and were replaced by CD127+ cells, indicating a progressive transition of effector cells to memory cells that do not originate from CD62L cells. The sub-types 3, 6 and 11 dominated during this recovery phase, expressing NF-kB, Jun/Fos, IFN-ɣ, TNF, and LT-⍺, and correspond to different genetic signatures (the genes TNF, IFIT2/3, MT-CO1/2, MT-ATP6). The specificity of viral antigens was less broad. There was progression from a T_EFF profile to an effector memory profile (T_EM). After a year, they all expressed CD45RA and less CCR7, characteristic of mature T_EMRA cells. In parallel, memory stem cells (T_SCM) and TCF1+ cells appeared. For reasons which are not yet clear, the T_EMRA cells are dominant in severe cases, while T_SCM dominate in moderate cases.  

This study therefore helps our understanding of how our immunity maintains a balance between effector T cells and T memory cells when infection by SARS-CoV-2 occurs. The majority of T-CD8 effector cells proliferate abundantly at the beginning of infection, but ultimately disappear. But certain sub-populations, where the signature has been indicated, become memory T cells which circulate over the long term. However, these observations need to be confirmed in tissues and for patients expressing different MHC profiles.