Differential regulatory T cell signature after recovery from mild COVID-19

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is characterized by a range of symptoms in which host immune response have been associated with disease progression. However, the putative role of regulatory T cells (Tregs) in determining COVID-19 outcomes has not been thoroughly investigated. Here, we compared peripheral Tregs between volunteers not previously infected with SARS-CoV-2 (healthy control [HC]) and volunteers who recovered from mild (Mild Recovered) and severe (Severe Recovered) COVID-19. Peripheral blood mononuclear cells (PBMC) were stimulated with SARS-CoV-2 synthetic peptides (Pool Spike CoV-2 and Pool CoV-2) or staphylococcal enterotoxin B (SEB). Results of a multicolor flow cytometric assay showed higher Treg frequency and expression of IL-10, IL-17, perforin, granzyme B, PD-1, and CD39/CD73 co-expression in Treg among the PBMC from the Mild Recovered group than in the Severe Recovered or HC groups for certain SARS-CoV-2 related stimulus. Moreover, Mild Recovered unstimulated samples presented a higher Tregs frequency and expression of IL-10 and granzyme B than did that of HC. Compared with Pool CoV-2 stimuli, Pool Spike CoV-2 reduced IL-10 expression and improved PD-1 expression in Tregs from volunteers in the Mild Recovered group. Interestingly, Pool Spike CoV-2 elicited a decrease in Treg IL-17+ frequency in the Severe Recovered group. In HC, the expression of latency-associated peptide (LAP) and cytotoxic granule co-expression by Tregs was higher in Pool CoV-2 stimulated samples. While Pool Spike CoV-2 stimulation reduced the frequency of IL-10+ and CTLA-4+ Tregs in PBMC from volunteers in the Mild Recovered group who had not experienced certain symptoms, higher levels of perforin and perforin+granzyme B+ co-expression by Tregs were found in the Mild Recovered group in volunteers who had experienced dyspnea. Finally, we found differential expression of CD39 and CD73 among volunteers in the Mild Recovered group between those who had and had not experienced musculoskeletal pain. Collectively, our study suggests that changes in the immunosuppressive repertoire of Tregs can influence the development of a distinct COVID-19 clinical profile, revealing that a possible modulation of Tregs exists among volunteers of the Mild Recovered group between those who did and did not develop certain symptoms, leading to mild disease.

with SARS-CoV-2 infection (Healthy Control -HC, n = 8) and volunteers who had recovered from mild (Mild Recovered, n = 9) and severe (Severe Recovered, n = 7) COVID-19. (A) The strategy of analysis for regulatory T cell identification and counting. The representative dot plots were derived from unstimulated samples. (B) Heatmap of frequency of regulatory T cells in volunteers from the HC, Mild Recovered, and Severe Recovered groups based on flow cytometry analyses. Each row indicates the condition in which the PBMCs were submitted (shown on the right). Each column indicates a volunteer enrolled in the study. (C.1-8) Comparison of the frequency of regulatory T cells among Mild Recovered volunteers who had and had not experienced the following symptoms: dyspnea, myalgia, sore throat, arthralgia, fatigue, diarrhea, anosmia, and ageusia during acute COVID-19.
The symbols •, ■, and ▲ represent each sample from Mild Recovered volunteer that was unstimulated (medium condition only) or stimulated with Pool Spike CoV-2 peptides or Pool CoV-2 peptides, respectively. When appropriate, multiple comparisons were performed using one-way ANOVA and Tukey's post hoc test or Kruskal-Wallis with Dunn's multiple comparisons tests. The bars represent mean values, and the error bars show the standard error of the mean (SEM) for each group.
Supplementary Figure 2. Regulatory T cells expressing IL-10, IL-17, and latency-associated peptide (LAP) in peripheral blood from volunteers. Peripheral blood mononuclear cells (PBMC) were collected from volunteers not previously affected by SARS-CoV-2 infection (Healthy Control -HC, n = 8) and volunteers who recovered from mild (Mild Recovered, n = 9) and severe (Severe Recovered, n = 7) COVID-19. (A) The strategy of analysis for regulatory T cells expressing IL-10, IL-17, and LAP identification and counting. The representative dot plots were derived from unstimulated samples. (B) Heatmap of frequency of regulatory T cells expressing IL-10, IL-17, and LAP in volunteers from the HC, Mild Recovered, and Severe Recovered groups based on flow cytometry analyses. Each row indicates the condition in which the PBMCs were submitted (shown on the right). Each column indicates a volunteer enrolled in the study. Each sequence of 4 rows indicates a marker analyzed (shown on the left). (C.1-3) Comparison of the frequency of regulatory T cells expressing IL-10 among Mild Recovered volunteers who had and had not experienced the following symptoms: dyspnea, sore throat, and ageusia during acute COVID-19. (C.4-9) Comparison of the frequency of regulatory T cells expressing IL-17 among Mild Recovered volunteers who had and had not experienced the following symptoms: dyspnea, sore throat, arthralgia, nasal obstruction, diarrhea, and ageusia during acute COVID-19. (C.10-15) Comparison of the frequency of regulatory T cells expressing LAP among Mild Recovered volunteers who had and had not experienced the following symptoms: myalgia, arthralgia, fatigue, diarrhea, anosmia, and ageusia during acute COVID-19.
The symbols •, ■, and ▲ represent each sample from Mild Recovered volunteer that was unstimulated (medium condition only) or stimulated with Pool Spike CoV-2 peptides or Pool CoV-2 peptides, respectively. When appropriate, multiple comparisons were performed using one-way ANOVA and Tukey's post hoc test or Kruskal-Wallis with Dunn's multiple comparisons tests. The bars represent mean values, and the error bars show the standard error of the mean (SEM) for each group.  .1-7) Comparison of the frequency of regulatory T cells expressing perforin among Mild Recovered volunteers who had and had not experienced the following symptoms: myalgia, sore throat, arthralgia, fatigue, nasal obstruction, diarrhea, and anosmia during acute COVID-19. (C.8-15) Comparison of the frequency of regulatory T cells expressing granzyme B among Mild Recovered volunteers who had and had not experienced the following symptoms: dyspnea, myalgia, sore throat, fatigue, nasal obstruction, diarrhea, anosmia, and ageusia during acute COVID-19. (C.16-21) Comparison of the frequency of regulatory T cells coexpressing perforin and granzyme B among Mild Recovered volunteers who had and had not experienced the following symptoms: myalgia, sore throat, fatigue, diarrhea, anosmia, and ageusia during acute COVID-19.
The symbols •, ■, and ▲ represent each sample from Mild Recovered volunteer that was unstimulated (medium condition only) or stimulated with Pool Spike CoV-2 peptides or Pool CoV-2 peptides, respectively. When appropriate, multiple comparisons were performed using one-way ANOVA and Tukey's post hoc test or Kruskal-Wallis with Dunn's multiple comparisons tests. The bars represent mean values, and the error bars show the standard error of the mean (SEM) for each group. The strategy of analysis for regulatory T cells expressing CD39 and CD73 and coexpressing CD39/CD73 identification and counting. The representative dot plots were derived from unstimulated samples. (B) Heatmap of frequency of regulatory T cells expressing CD39, CD73, and coexpressing both ectonucleotidases in volunteers from the HC, Mild Recovered, and Severe Recovered groups based on flow cytometry analyses. Each row indicates the condition in which the PBMCs were submitted (shown on the right). Each column indicates a volunteer enrolled in the study. Each sequence of 4 rows indicates a marker analyzed (shown on the left). (C.1-5) Comparison of the frequency of regulatory T cells expressing CD39 among Mild Recovered volunteers who had and had not experienced the following symptoms: dyspnea, sore throat, nasal obstruction, diarrhea, and ageusia during acute COVID-19. (C.6-11) Comparison of the frequency of regulatory T cells expressing CD73 among Mild Recovered volunteers who had and had not experienced the following symptoms: dyspnea, sore throat, fatigue, nasal obstruction, anosmia, and ageusia during acute COVID-19. (C.12-18) Comparison of the frequency of regulatory T cells coexpressing CD39 and CD73 among Mild Recovered volunteers who had and had not experienced the following symptoms: dyspnea, sore throat, arthralgia, fatigue, nasal obstruction, anosmia, and ageusia during acute COVID-19.

Supplementary
The symbols •, ■, and ▲ represent each sample from Mild Recovered volunteer that was unstimulated (medium condition only) or stimulated with Pool Spike CoV-2 peptides or Pool CoV-2 peptides, respectively. When appropriate, multiple comparisons were performed using one-way ANOVA and Tukey's post hoc test or Kruskal-Wallis with Dunn's multiple comparisons tests. The bars represent mean values, and the error bars show the standard error of the mean (SEM) for each group.