Relationship of SARS-CoV-2-specific CD4 response to COVID-19 severity and impact of HIV-1 and tuberculosis coinfection

Abstract

T cells are involved in control of coronavirus disease 2019 (COVID-19), but limited knowledge is available on the relationship between antigen-specific T cell response and disease severity. Here, we used flow cytometry to assess the magnitude, function, and phenotype of SARS coronavirus 2-specific (SARS-CoV-2-specific) CD4+ T cells in 95 hospitalized COVID-19 patients, 38 of them being HIV-1 and/or tuberculosis (TB) coinfected, and 38 non-COVID-19 patients. We showed that SARS-CoV-2-specific CD4+ T cell attributes, rather than magnitude, were associated with disease severity, with severe disease being characterized by poor polyfunctional potential, reduced proliferation capacity, and enhanced HLA-DR expression. Moreover, HIV-1 and TB coinfection skewed the SARS-CoV-2 T cell response. HIV-1-mediated CD4+ T cell depletion associated with suboptimal T cell and humoral immune responses to SARS-CoV-2, and a decrease in the polyfunctional capacity of SARS-CoV-2-specific CD4+ T cells was observed in COVID-19 patients with active TB. Our results also revealed that COVID-19 patients displayed reduced frequency of Mycobacterium tuberculosis-specific CD4+ T cells, with possible implications for TB disease progression. These results corroborate the important role of SARS-CoV-2-specific T cells in COVID-19 pathogenesis and support the concept of altered T cell functions in patients with severe disease.

Keywords: AIDS/HIV; COVID-19; Cellular immune response; T cells; Tuberculosis.

Figure 1. Measures of COVID-19 disease severity.

(A) An unsupervised 2-way hierarchical cluster analysis (HCA, Ward’s method) was employed to grade COVID-19 disease, using the WHO ordinal scale scoring, Roche Elecsys anti–SARS-CoV-2 antibody cutoff index, WCC, CRP, D-dimer, ferritin, LDH, and radiographic evidence of disease extent expressed as percentage of unaffected lung. COVID-19 status (COVID-19 cases in red and SARS-CoV-2–uninfected hospitalized controls in blue) and outcome (survived in white and deceased in black) of each patient is indicated at the top of the dendrogram. Data are depicted as a heatmap colored from minimum to maximum values detected for each parameter. (B) Constellation plot-cluster analysis based on all measured parameters. Each dot represents a participant and is color-coded according to his or her COVID-19 status. Each cluster obtained for the HCA is identified by a number. (C) Principal component analysis (PCA) on correlations, based on the 8 clinical parameters, was used to explain the variance of the data distribution in the cohort. Each dot represents a participant. The 2 axes represent principal components 1 (PC1) and 2 (PC2). Their contribution to the total data variance is shown as a percentage. (D) Loading plot showing how each parameter influences PC1 and PC2 values. (E) Comparison of PC1 score values between COVID-19 cases who survived and those who died. Bars represent medians. Statistical comparisons were calculated using the nonparametric Mann-Whitney U test. Only participants with complete clinical data were included in the analysis (n = 79 COVID-19 patients and n = 25 hospitalized controls).

Figure 2. Prevalence, magnitude, and functional profile of SARS-CoV-2–specific CD4⁺ T cells between COVID-19 cases and SARS-CoV-2–uninfected hospitalized patients.

(A) Representative flow cytometry plots of IFN-γ and TNF-α expression. NS, no stimulation. (B) Proportion of patients exhibiting a detectable SARS-CoV-2 CD4 response in each group. The number of studied patients is indicated in the pie (n = 79 COVID-19 patients and n = 25 hospitalized controls). (C) Frequency of SARS-CoV-2–specific CD4⁺ T cells in hospitalized control (blue, n = 13) and COVID-19 responders (red, n = 79). Statistical comparisons were calculated using the nonparametric Mann-Whitney U test. (D) Polyfunctional profile of SARS-CoV-2–specific CD4⁺ T cells in hospitalized controls and COVID-19 patients. The median and IQR are shown. Each response pattern is color-coded, and data are summarized in the pie charts. Wilcoxon’s rank test was used to compare response patterns between groups. Statistical differences between pies were defined using a permutation test.

Figure 3. Memory and activation profile of SARS-CoV-2–specific CD4⁺ T cells between COVID-19 cases and SARS-CoV-2–uninfected hospitalized patients.

(A) Overlay flow plots of CD45RA, CD27, PD-1, GrB, CD38, and HLA-DR expression. Dots depict SARS-CoV-2–specific CD4⁺ T cells and density plots depict total CD4⁺ T cells. Four memory subsets can be delineated: naive (CD45RA⁺CD27⁺), early differentiated (ED, CD45RA^–CD27⁺), late differentiated (LD, CD45RA^–CD27^–), and effector (Eff, CD45RA⁺CD27^–). (B) Summary graphs of the expression of each marker in SARS-CoV-2–specific CD4⁺ T cells (n = 75 COVID-19 patients and n = 12 hospitalized controls). The phenotype of SARS-CoV-2–specific CD4⁺ T cells was assessed only in those with response greater than 20 events. Bars represent medians. Statistical comparisons were calculated using the nonparametric Mann-Whitney U test. (C) Heatmap of pairwise Spearman’s correlations between phenotypical and functional traits of SARS-CoV-2–specific CD4⁺ T cells. Spearman’s rank r correlation values are shown from blue, –1, to yellow, 1. The red box identifies the profile of ED SARS-CoV-2–specific CD4⁺ T cells and the blue box the profile of LD cells enriched in hospitalized controls. (D) PCA (left) based on the 8 phenotypical and functional attributes of SARS-CoV-2–specific CD4⁺ T cells (LD, GrB, HLA-DR, Ki67, CD38 and the proportion of IFN-γ⁺IL-2⁺TNF-α⁺, IFN-γ⁺IL-2^–TNF-α⁺, and IFN-γ^–IL-2^–TNF-α⁺ cells) and corresponding loading plot (right).

Figure 4. SARS-CoV-2–specific CD4⁺ T cell response in COVID-19 cases stratified by WHO ordinal scale score and outcome.

(A) Prevalence and frequency of SARS-CoV-2–specific CD4⁺ T cells in COVID-19 cases. Patients were stratified according to WHO ordinal score and outcome. (B) Polyfunctional profile of SARS-CoV-2–specific CD4⁺ T cells in COVID-19 cases stratified by WHO score and outcome. Wilcoxon’s rank test was used to compare response patterns between groups (*P < 0.05, **P < 0.01, ***P < 0.001). Statistical differences between pie charts were defined using a permutation test. (C) Memory and activation profile of SARS-CoV-2–specific CD4⁺ T cells in COVID-19 cases stratified by WHO score and outcome. The phenotype of SARS-CoV-2–specific CD4⁺ T cells was assessed only in those with response greater than 20 events (n = 75 COVID-19 patients). Statistical comparisons were defined using a Kruskal-Wallis test, adjusted for multiple comparisons (Dunn’s test) for the different WHO groups and the Mann-Whitney U test to compare COVID-19 patients who survived or died.

Figure 5. Relationship between COVID-19 severity and functional and phenotypical traits of SARS-CoV-2–specific CD4⁺ T cells.

(A) Spearman’s correlation r values between indicated SARS-CoV-2–specific CD4⁺ T cell features and COVID-19 severity (defined by the composite analysis of clinical parameters, PC1 severity). Negative associations are represented in blue and positive associations in yellow. P values are indicated for each comparison. (B) Comparison of the overall profile of SARS-CoV-2–specific CD4⁺ T cells (PC2 phenotype) in COVID-19 cases (n = 74) stratified by WHO ordinal score and outcome. Statistical comparisons were defined using a Kruskal-Wallis test, adjusted for multiple comparisons (Dunn’s test) for the different WHO groups and the Mann-Whitney U test to compare COVID-19 patients who survived or died. (C) Association between COVID-19 severity (PC1 severity) and the overall profile of SARS-CoV-2–specific CD4⁺ T cells (PC2 phenotype). COVID-19 survivors are depicted in gray and patients who died in black. Correlation was tested by a 2-tailed nonparametric Spearman’s rank test.

Figure 6. Impact of HIV, aTB, and HIV/aTB coinfection on SARS-CoV-2–specific CD4⁺ T cell response.

(A) Comparison of COVID-19 severity (defined by the composite analysis of clinical parameters, PC1 severity) between patients grouped according to HIV and/or aTB coinfection. (B) Prevalence and frequencies of SARS-CoV-2–specific CD4⁺ T cells in COVID-19 patients stratified by HIV and/or aTB coinfection. Statistical comparisons were defined using a Kruskal-Wallis test adjusted for multiple comparisons (Dunn’s test). (C) Comparison of the frequency of total CD4⁺ T cells between SARS-CoV-2 CD4 responders and nonresponders. Dots are color-coded according to patient’s HIV and TB status. Statistical comparison was performed using the Mann-Whitney U test. (D) Association between the frequency of SARS-CoV-2–specific CD4⁺ T cells and total CD4⁺ T cells in HIV-infected COVID-19 patients. Correlation was tested by a 2-tailed nonparametric Spearman’s rank test. (E) Prevalence and magnitude of SARS-CoV-2–specific serological response (defined using the Roche Elecsys assay) in COVID-19 patients stratified by HIV and/or aTB coinfection. (F) Association between the magnitude of SARS-CoV-2–specific serological response and the frequency of total CD4⁺ T cells in HIV-infected COVID-19 patients. Correlation was tested by a 2-tailed nonparametric Spearman’s rank test. (G) Polyfunctional profile of SARS-CoV-2–specific CD4⁺ T cells in COVID-19 cases stratified by HIV or aTB coinfection. For this analysis, HIV^–/aTB⁺ and HIV⁺/aTB⁺ patients were combined in 1 group (aTB). Dots are color-coded according to patients’ HIV and TB status. Wilcoxon’s rank test was used to compare response patterns between groups (**P < 0.01). Statistical differences between pie charts were defined using a permutation test. (H) Comparison of the overall profile of SARS-CoV-2–specific CD4⁺ T cells (PC2 phenotype) in COVID-19 cases stratified by HIV or aTB coinfection. Statistical comparisons were defined using a Kruskal-Wallis test adjusted for multiple comparisons (Dunn’s test).

Figure 7. Impact of COVID-19 on M.

tuberculosis–specific CD4⁺ T cell response. (A) Representative examples of flow cytometry plots of SARS-CoV-2– and M. tuberculosis–specific CD4⁺ T cell responses in 3 COVID-19 patients (1 HIV^–/aTB^–, 1 HIV⁺/aTB^–, and 1 HIV⁺/aTB⁺). (B) Comparison of the prevalence and frequencies of SARS-CoV-2– and M. tuberculosis–specific CD4⁺ T cells in COVID-19 patients stratified by HIV or aTB coinfection. The proportion of responders to each pathogen (S: SARS-CoV-2 and M: M. tuberculosis) is presented with pies at the top of the graph. Statistical comparisons were performed using the χ² test. Participants were grouped according to their HIV and/or TB status. Black bars represent the medians. (C) Comparisons of the frequencies of M. tuberculosis–specific CD4⁺ T cells in a cohort recruited before the emergence of COVID-19 (2018, n = 114), SARS-CoV-2–uninfected hospitalized controls (n = 29), and COVID-19 cases (n = 76). Participants were stratified according to their HIV and/or TB status. Statistical comparisons were defined using a Kruskal-Wallis test adjusted for multiple comparisons (Dunn’s test) for each subgroup. (D) Representative flow cytometry plots of HLA-DR expression on TNF-α–producing M. tuberculosis–specific CD4⁺ T cells in 3 COVID-19 patients (1 HIV^–/aTB^–, 1 HIV⁺/aTB^–, and 1 HIV⁺/aTB⁺). (E) Summary graph of HLA-DR expression on M. tuberculosis–specific CD4⁺ T cells in a cohort recruited before the emergence of COVID-19 (2018), SARS-CoV-2–uninfected hospitalized controls, and COVID-19 cases stratified according to HIV and TB status. The phenotype of M. tuberculosis–specific CD4⁺ T cells was assessed only in those with response greater than 20 events.