CD4+ T-Cell Dysfunction in Severe COVID-19 Disease Is Tumor Necrosis Factor-α/Tumor Necrosis Factor Receptor 1-Dependent

Abstract

Rationale: Lymphopenia is common in severe coronavirus disease (COVID-19), yet the immune mechanisms are poorly understood. As inflammatory cytokines are increased in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, we hypothesized a role in contributing to reduced T-cell numbers. Objectives: We sought to characterize the functional SARS-CoV-2 T-cell responses in patients with severe versus recovered, mild COVID-19 to determine whether differences were detectable. Methods: Using flow cytometry and single-cell RNA sequence analyses, we assessed SARS-CoV-2-specific responses in our cohort. Measurements and Main Results: In 148 patients with severe COVID-19, we found lymphopenia was associated with worse survival. CD4⁺ lymphopenia predominated, with lower CD4⁺/CD8⁺ ratios in severe COVID-19 compared with patients with mild disease (P < 0.0001). In severe disease, immunodominant CD4⁺ T-cell responses to Spike-1 (S1) produced increased in vitro TNF-α (tumor necrosis factor-α) but demonstrated impaired S1-specific proliferation and increased susceptibility to activation-induced cell death after antigen exposure. CD4⁺TNF-α⁺ T-cell responses inversely correlated with absolute CD4⁺ counts from patients with severe COVID-19 (n = 76; R = -0.797; P < 0.0001). In vitro TNF-α blockade, including infliximab or anti-TNF receptor 1 antibodies, strikingly rescued S1-specific CD4⁺ T-cell proliferation and abrogated S1-specific activation-induced cell death in peripheral blood mononuclear cells from patients with severe COVID-19 (P < 0.001). Single-cell RNA sequencing demonstrated marked downregulation of type-1 cytokines and NFκB signaling in S1-stimulated CD4⁺ cells with infliximab treatment. We also evaluated BAL and lung explant CD4⁺ T cells recovered from patients with severe COVID-19 and observed that lung T cells produced higher TNF-α compared with peripheral blood mononuclear cells. Conclusions: Together, our findings show CD4⁺ dysfunction in severe COVID-19 is TNF-α/TNF receptor 1-dependent through immune mechanisms that may contribute to lymphopenia. TNF-α blockade may be beneficial in severe COVID-19.

Keywords: CD4+ T cells; COVID-19; SARS-CoV-2 infection; TNF-α; lymphopenia.

Figure 1.

Peripheral lymphopenia is associated with mortality and a predominant CD4⁺ T-cell lymphopenia in severe coronavirus disease (COVID-19). (A) The distribution of absolute lymphocyte counts (ALC) for n = 148 patients with severe COVID-19. Values represent the first ALC obtained upon admission and the median ALC (red line) was 700 cells/mm³. (B) Kaplan-Meier 30-day survival curves for patients with severe COVID-19 with initial ALC values at or above the median (ALC ⩾ 700) (n = 78) versus those median (ALC <700) (n = 70), showing a significant difference in survival between groups (log-rank test, P = 0.012). (C) Kaplan-Meier 30-day survival curves on (n = 76) patients with severe COVID-19 with initial CD4⁺ lymphocyte values ⩾200 (n = 31) versus those <200 (n = 45), showing a borderline difference in survival between groups (log-rank test, P = 0.053). (D) Representative flow cytometry plot of CD4⁺ and CD8⁺ T cells from patients with severe COVID-19 with lymphopenia ALC <700 (left panel) and nonlymphopenia ALC ⩾700 (right panel). (E–H) Cumulative data (n = 76) from patients with severe COVID-19 with lymphopenia (ALC <700) (n = 38) (solid red squares) compared with (E) nonlymphopenia (ALC ⩾ 700) (n = 38) (open red squares) showing frequencies of CD4⁺ and CD8⁺ T-cell subsets, (F) CD4⁺/CD8⁺ T-cell ratios, (G) the absolute CD4⁺ T-cells number (cells/mm³), (H) and CD8⁺ T cell numbers (cells/mm³). (I) Cumulative data showing CD4⁺/CD8⁺ T-cell ratio from the patient cohort used in functional studies (n = 48) with mild disease (n = 24) (blue squares) compared with severe disease (n = 24) (red squares) and with healthy donors (n = 15) (green squares). (J) Representative flow cytometry plots of CD4⁺and CD8⁺ T-cell frequencies from patients with mild (left panel) and severe COVID-19 (middle panel) and a healthy donor (right panel).

Figure 2.

CD4⁺TNF-α⁺ (tumor necrosis factor-α⁺) and CD4⁺CD107a⁺ Spike-1 (S1)-specific effector responses are immunodominant and increased in patients with severe coronavirus disease (COVID-19). (A and B) Pooled data of patients with intracellular cytokine staining (ICS) immunodominance (n = 48) used in our T-cell immunity functional cohort studies for COVID-19–specific S1, Spike 2 (S2), viral envelope small membrane protein (VEMP), nucleocapsid (NCAP), and staphylococcal enterotoxin B (SEB)-reactive of CD4⁺IFNγ⁺ (A) and CD4⁺TNFα⁺ (B) effector responses. (C) Representative flow cytometry plots of ICS of severe patient CD4⁺TNFα⁺ responses for medium, COVID-19–specific antigens S1, S2 (upper panels), VEMP, NCAP, and positive control–SEB (lower panels). (D) Cumulative data of ICS of S1-specific COVID-19 CD4⁺ (black columns) and CD8⁺ (gray columns) of T-cell effector responses (IFNγ, TNFα, CD107a, IL-2, IL-13, and IL-17) in (n = 48) patients used in functional studies. (E and F) Pooled data showing ICS of S1-specific COVID-19 CD4⁺ (E) and CD8⁺ (F) T-cell effector responses of IFNγ, TNFα, CD107a, IL-2, IL-13, and IL-17 in mild (blue columns) (n = 24) versus those with severe (red columns) (n = 24) COVID-19 disease. (G and H) Individual pie charts showing S1-specific multifunctional responses in severe COVID-19 of CD4⁺ (left pie) and CD8⁺ (right pie) (n = 24) (G) and CD4⁺ in severe (left pie) (n = 24) versus CD4⁺ in mild COVID-19 disease (right pie) (n = 24) (H). The T cell multifunctional responses were evaluated for four cytokines: TNFα (red arch), IFNγ (green arch), IL-2 (yellow arch), and CD107a (blue arch), and the combination of cytokine responses were: +4, yellow pie fraction; +3, green pie fraction; +2, blue pie fraction; and +1, red pie fraction. (I) Inverse correlation of S1-specific COVID-19 CD4⁺TNFα⁺ response frequencies with CD4⁺/CD8⁺ ratio on patients with severe (red dots) and mild (blue dots) COVID-19 (n = 48). (J) The inverse correlation of S1-specific COVID-19 CD4⁺TNFα⁺ response frequencies with absolute CD4⁺ number (cells/mm³) on severe COVID-19 (n = 76) (red dots).

Figure 3.

Impaired Spike-1 (S1)-specific CD4⁺ T-cell proliferation in severe coronavirus disease (COVID-19) correlates with lymphopenia, is TNF-α (tumor necrosis factor-α)/TNFRI (TNF receptor 1)-dependent, and can be rescued in vitro by infliximab. (A and B) Representative flow cytometric plots (A) and cumulative data (B) showing Day 6 S1-specific CD4⁺ T-cell proliferation by carboxyfluorescein succinimidyl ester (CFSE) dilution from patients with mild COVID-19 (left panel [A] and blue square [B]) versus severe (right panel [A] and red square [B]). (C and D) Direct correlation of S1-specific COVID-19 CD4⁺ T-cell proliferation (Day 6) from mild (blue dots) and severe COVID-19 (red dots) (n = 48), response frequencies with CD4⁺/CD8⁺ ratio (C), and with absolute CD4⁺ numbers in patients with severe COVID-19 with lymphopenia (open red dots) and nonlymphopenia (solid red dots) (n = 24) (D). (E and F) Representative flow cytometric plots (E) and cumulative data (F) showing S1-specific CD4⁺ T-cell proliferation from CD8⁺-depleted peripheral blood mononuclear cells (PBMC) in patients with severe disease (n = 12) (red squares) in the presence or absence of anti–TNF-α antibodies. (G and H) Representative histogram plots (G) and cumulative data (H) showing CD4⁺ T-cell TNFRI⁺ (CD120a) (left panel [G]) and TNFRII⁺ (CD120b) (right panel [G]) surface expression in mild (n = 24) COVID-19 (blue lines [G] and blue dots [H]) overlayed with severe COVID-19 (red lines [G] and red dots [H]) (n = 24) and normal healthy donors (green lines [G] and green dots [H]) (n = 15) and the isotype control (full gray lines [G]), and the numbers represent the percentage of positive expression of TNFRI and TNFRII. (I and J) Representative flow cytometric plots (I) and cumulative data (J) of CD4⁺ S1-specific proliferation in the presence or absence of anti-TNFRI or anti-TNFRII antibodies in CD8⁺-depleted PBMC from patients with severe COVID-19 (red squares) (n = 12). (K–M) Representative flow cytometric plots (K) and cumulative data (L) showing CD4⁺ S1-specific proliferation in the presence or absence of infliximab at various doses and the presence or absence of anti-PD1 antibodies (M) and isotype control (IgG1) (red squares). (N and O) Representative flow cytometric plots (N) and cumulative data (O) showing S1-specific frequencies of CD4⁺ T-cell effector responses (IFNγ, TNFα, CD107a, and IL-2) after 6 h restimulation with S1 peptides and infliximab (lower panel [N], green columns [O]) or without infliximab treatment (upper panels [N], orange columns [O]) for patients with severe COVID-19 (n = 12).

Figure 3.

Impaired Spike-1 (S1)-specific CD4⁺ T-cell proliferation in severe coronavirus disease (COVID-19) correlates with lymphopenia, is TNF-α (tumor necrosis factor-α)/TNFRI (TNF receptor 1)-dependent, and can be rescued in vitro by infliximab. (A and B) Representative flow cytometric plots (A) and cumulative data (B) showing Day 6 S1-specific CD4⁺ T-cell proliferation by carboxyfluorescein succinimidyl ester (CFSE) dilution from patients with mild COVID-19 (left panel [A] and blue square [B]) versus severe (right panel [A] and red square [B]). (C and D) Direct correlation of S1-specific COVID-19 CD4⁺ T-cell proliferation (Day 6) from mild (blue dots) and severe COVID-19 (red dots) (n = 48), response frequencies with CD4⁺/CD8⁺ ratio (C), and with absolute CD4⁺ numbers in patients with severe COVID-19 with lymphopenia (open red dots) and nonlymphopenia (solid red dots) (n = 24) (D). (E and F) Representative flow cytometric plots (E) and cumulative data (F) showing S1-specific CD4⁺ T-cell proliferation from CD8⁺-depleted peripheral blood mononuclear cells (PBMC) in patients with severe disease (n = 12) (red squares) in the presence or absence of anti–TNF-α antibodies. (G and H) Representative histogram plots (G) and cumulative data (H) showing CD4⁺ T-cell TNFRI⁺ (CD120a) (left panel [G]) and TNFRII⁺ (CD120b) (right panel [G]) surface expression in mild (n = 24) COVID-19 (blue lines [G] and blue dots [H]) overlayed with severe COVID-19 (red lines [G] and red dots [H]) (n = 24) and normal healthy donors (green lines [G] and green dots [H]) (n = 15) and the isotype control (full gray lines [G]), and the numbers represent the percentage of positive expression of TNFRI and TNFRII. (I and J) Representative flow cytometric plots (I) and cumulative data (J) of CD4⁺ S1-specific proliferation in the presence or absence of anti-TNFRI or anti-TNFRII antibodies in CD8⁺-depleted PBMC from patients with severe COVID-19 (red squares) (n = 12). (K–M) Representative flow cytometric plots (K) and cumulative data (L) showing CD4⁺ S1-specific proliferation in the presence or absence of infliximab at various doses and the presence or absence of anti-PD1 antibodies (M) and isotype control (IgG1) (red squares). (N and O) Representative flow cytometric plots (N) and cumulative data (O) showing S1-specific frequencies of CD4⁺ T-cell effector responses (IFNγ, TNFα, CD107a, and IL-2) after 6 h restimulation with S1 peptides and infliximab (lower panel [N], green columns [O]) or without infliximab treatment (upper panels [N], orange columns [O]) for patients with severe COVID-19 (n = 12).

Figure 4.

Spike-1 (S1) induces activation-induced cell death in CD4⁺ T cells from patients with severe disease and is TNF-α (tumor necrosis factor-α)/TNFRI (TNF receptor 1)-dependent. (A and B) Representative histograms overlayed from patients with severe (red line) versus mild (blue line) coronavirus disease (COVID-19) and medium as control (gray full line) (A) and pooled data depicting the annexin V⁺ staining in CD4⁺ T cells in patients with severe (n = 12) (red column) versus mild (n = 12) (blue column) COVID-19 versus medium (gray column) (B). (C and D) Representative histograms (C) and pooled data (D) showing the CD4⁺annexin V⁺ in CD8⁺-depleted peripheral blood mononuclear cells from patients with severe COVID-19 (n = 12). Cells were cultured in the presence or absence of anti–TNF-α neutralizing antibodies (dark blue histogram lines [C] and dark blue column [D]), infliximab at various doses (green, black, and light blue histogram lines [C] and columns [D]), anti-TNFRI antibodies (purple histogram lines [C] and purple column [D]) or anti-TNFRII (yellow histogram lines [C] and yellow column [D]) antibodies, compared with S1-specific isotype control (IgG1) (red histogram lines [C] and red column [D]) or medium alone (gray full histogram lines [C] and gray column [D]). (E and F) Representative histograms (E) and pooled data (F) showing the CD4⁺annexin V⁺ from patients with severe COVID-19 in the presence or absence of anti-Fas neutralizing antibodies (black histogram lines [E] and black column [F]), compared with anti–TNF-α neutralizing antibodies (dark blue histogram lines [E] and dark blue column [F]) versus S1-specific isotype control (IgG1) (red histogram lines [E] and red column [F]) and medium as control (gray full line [E] and gray column [F]). (G and H) Representative histograms (G) and pooled data (H) depicting the CD4⁺annexin V⁺ from patients with severe COVID-19 in the presence or absence of anti–TNF-related apoptosis-inducing ligand (TRAIL) neutralizing antibodies at various doses (black and purple histogram lines [G] and black and purple columns [H]), compared with anti–TNF-α neutralizing antibodies (dark blue histogram lines [G] and dark blue column [H]), versus S1-specific isotype control (IgG1) (red histogram lines [G] and red column [H]) and medium as control (gray full line [G] and gray column [H]).

Figure 5.

High TNF-α (tumor necrosis factor-α) production by lung resident memory CD4⁺ T cells in patients with severe coronavirus disease (COVID-19). (A) Representative flow cytometry plots on explanted lung tissue in patients with severe COVID-19 of CD4⁺ and CD8⁺ T-cell frequencies in peripheral blood mononuclear cells (PBMC) (upper panel); lung parenchyma (LP) (middle panel), and BAL fluid–derived cells (lower panel). (B) Representative flow cytometry plots of the same patient with COVID-19 showing CD4⁺ T-cell frequencies at 6 h after Spike-1 (S1)-specific peptides stimulation for TNF-α, IFN-γ, IL-2, and CD107a in PBMC (upper panels); LP (middle panels), and BAL fluid–derived cells (lower panels). (C) Pooled data showing intracellular staining of 6 h S1-specific CD4⁺ T-cell effector responses: TNF-α, IFN-γ, IL-2, and CD107a in BAL fluid-derived cells from n = 6 as lung transplant recipients (dark blue columns) versus PBMC (dark red columns). (D) Representative flow cytometry plots of the COVID-19 lung explant showing CD4⁺ T-cell surface expression of CD45RA⁺, CCR7⁺ (left panel), and CD69⁺CD103⁺ (right panel) on PBMC (red) overlayed with LP (blue) lung; values represent cell frequencies of T-cell subsets. (E) Flow cytometry histograms of the COVID-19 explant showing LP lung CD4⁺ T cells CD38⁺ (left panel), PD1⁺ (middle panel), and Ki67⁺ (right panel) overlayed with LP lung (blue lines) on PBMC (full pink lines).