Upon Ag encounter, T cells can rapidly divide and form an effector population, which plays an important role in fighting acute infections. In humans, little is known about the molecular markers that distinguish such effector cells from other T cell populations. To address this, we investigated the molecular profile of T cells present in individuals with active tuberculosis (ATB), where we expect Ag encounter and expansion of effector cells to occur at higher frequency in contrast to Mycobacterium tuberculosis–sensitized healthy IGRA+ individuals. We found that the frequency of HLA-DR+ cells was increased in circulating CD4 T cells of ATB patients, and was dominantly expressed in M. tuberculosis Ag–specific CD4 T cells. We tested and confirmed that HLA-DR is a marker of recently divided CD4 T cells upon M. tuberculosis Ag exposure using an in vitro model examining the response of resting memory T cells from healthy IGRA+ to Ags. Thus, HLA-DR marks a CD4 T cell population that can be directly detected ex vivo in human peripheral blood, whose frequency is increased during ATB disease and contains recently divided Ag-specific effector T cells. These findings will facilitate the monitoring and study of disease-specific effector T cell responses in the context of ATB and other infections.

This article is featured in Top Reads, p.357

Effector T cells originate from division of Ag-specific T cells upon TCR activation (1). They can be derived from either naive T cells upon primary exposure with Ag, or from memory T cells upon Ag re-exposure (2). Effector T cells show higher levels of differentiation and activation of effector functions, such as cytotoxicity and production of cytokines/chemokines, compared with memory T cells (35). The effector function of CD4 T cells is highly heterogeneous. It is predominantly mediated through the production of various cytokines and chemokines, and the expression of cell surface proteins to act on other immune cells, infected cells, or pathogens (6). CD4 T cells can also display cytotoxic functions and directly eliminate pathogens or infected cells (7, 8). By virtue of their heterogeneity, effector CD4 T cells are critical in controlling a wide range of pathogenic infections including bacteria, viruses, parasites, and fungi (69).

The ability to identify and isolate effector T cells in peripheral blood in the context of human infection is of high significance for several reasons. First, the effector population is enriched for Ag-specific T cells and is thus an ideal target for the study and isolation of Ag-specific T cells directly ex vivo. Second, examining the frequency and phenotype of effector cells in a given individual provides information on the immune system status and ongoing immune responses. In diagnosed active diseases such as tuberculosis (TB), the effector T cell population could be tracked to monitor the course of infection and response to treatment. In presymptomatic individuals, an increase in frequency of effector T cells could be used as a surrogate marker of increased Ag load, suggesting loss of immunological control of the pathogen. Finally, the ability to isolate effector T cells is necessary to shed light on their biology (e.g., generation, heterogeneity, and maintenance), which is still poorly understood, especially for human CD4 T cells.

Despite their importance, there is no single phenotypic marker that can identify bona fide effector CD4 T cells in humans. From parallel studies in the mouse model, effector human CD4 T cells have been reported to have a more “differentiated phenotype” with low expression levels of CCR7, CD27, CD28, CD62L, and CD127 and high expression levels of CD95, PD1, and KLRG1 (4, 1013). However, the combination of markers to define effector CD4 T cells tend to vary between studies and pathogens (14), and the resulting phenotype do overlap with other cell subsets such as effector memory T cells (10, 11). Thus, one key challenge to study effector T cells is the ability to readily distinguish, among circulating CD4 T cells of a given Ag specificity, those that are effector cells from those that are not.

In this study, we set out to characterize the phenotype of effector CD4 T cells in active TB (ATB). ATB is caused by infection with the bacterium Mycobacterium tuberculosis and is the leading cause of death from one single infectious agent worldwide, with ∼1.4 million deaths annually (15). Most infected individuals contain M. tuberculosis after initial infection, becoming latently infected, with no clinical symptoms and no risk of transmission. In some individuals, particularly those who are elderly or immunocompromised, M. tuberculosis can develop into active disease, which is associated with clinical symptoms, risk of transmission, and high mortality and morbidity (16, 17). This can occur either early postinfection, or after a period of latency (reactivation TB). CD4 T cells are important in the immune control of M. tuberculosis (18, 19), and there is evidence for the presence of M. tuberculosis–specific memory CD4 T cells in peripheral blood in all stages of M. tuberculosis infection (20, 21). However, we expect the prevalence of circulating M. tuberculosis–specific effector T cells to be increased in ATB, which is marked by a lack of control of M. tuberculosis and thus increased Ag encounter, leading to effector T cell formation and proliferation. Accordingly, we hypothesized that there is a higher frequency of effector CD4 T cells present in human peripheral blood of individuals with microbiologically confirmed ATB compared with M. tuberculosis–sensitized, healthy IGRA+ individuals. To test this, we compared the transcriptomic profile of circulating CD4 T cells of individuals with ATB to IGRA+ and IGRA individuals to give us insight into the molecular profile of human effector CD4 T cells.

Human study participants were enrolled at the University of California, San Diego Anti-Viral Research Center Clinic (United States), the Universidad Peruana Cayetano Heredia (Lima, Peru), the National Hospital for Respiratory Diseases, Welisara (Sri Lanka), or The South African Tuberculosis Vaccine Initiative, Western Cape Province (South Africa). Ethical approval to carry out this work was maintained through the La Jolla Institute for Immunology Institutional Review Board or the Human research Ethics Committee of the University of Cape Town. The University of Colombo Ethics Review Committee served as the National Institutes of Healthy registered Institutional Review Board for the Kotelawala Defense University. All clinical investigations were conducted according to the principles expressed in the Declaration of Helsinki, and all participants provided written informed consent prior to participation in the study. All samples were obtained for specific use in this study.

M. tuberculosis sensitization status was confirmed in participants by a positive IFN-γ–release assay (QuantiFERON-TB Gold In-Tube; Cellestis or T-SPOT.TB; Oxford Immunotec) and the absence of symptoms consistent with TB or other clinical or radiographic signs of ATB (healthy IGRA+ cohort). ATB was defined as 1) presence of clinical symptoms and/or radiological/histological evidence of pulmonary TB and 2) microbiologically confirmed by M. tuberculosis–specific molecular testing on sputum. Patients were not tested for the presence of nontuberculous mycobacteria infection. IGRA uninfected controls had no past medical history of TB, nor exposure to M. tuberculosis or evidence of M. tuberculosis sensitization as confirmed by a negative IFN-γ–release assay. A total of 39 ATB, 43 IGRA+, and 24 IGRA participants were analyzed in this study. All participants were confirmed negative for HIV. PBMCs were obtained by density gradient centrifugation (Ficoll-Hypaque, Amersham Biosciences) from leukapheresis or whole-blood samples, according to the manufacturer’s instructions. Cells were resuspended at 50–100 million cells per milliliter in FBS (Gemini Bio-Products) containing 10% DMSO (Sigma-Aldrich) and cryopreserved in liquid nitrogen.

Cryopreserved PBMC were quickly thawed by incubating each cryovial at 37°C for 2 min, and cells transferred into 9 ml of cold medium (RPMI 1640 with L-glutamin and 25 mM HEPES) (Omega Scientific), supplemented with 5% human AB serum (GemCell), 1% penicillin streptomycin (Life Technologies), and 1% glutamax (Life Technologies) and 20 U/ml benzonase nuclease (MilliporeSigma). Cells were centrifuged and resuspended in medium to determine cell concentration and viability using trypan blue and a hematocytometer. Cells were then kept at 4°C until use for flow cytometry, cell sorting, or Cell Trace Violet (CTV) staining. For the measurement of spontaneous cytokine production in HLA-DR+/ CD4 T cells by flow cytometry, prior to the staining, 0.5 × 106 cells were incubated with BD Golgi Plug (1:1000 dilution stock, BD Biosciences) for 2–3h at 37°C.

Flow cytometry experiments were performed as previously described (22, 23). For surface staining, up to 0.5 × 106 cells were incubated with 10% FBS in 1× PBS for 10 min. Cells were then stained with 100 μl of PBS containing 0.1 μl fixable viability dye eFluor506 (eBioscience, corresponding to 1:1000 dilution of the stock, as per the manufacturer’s recommendation), 2 μl of FcR blocking reagent (corresponding to 1:50 dilution of the stock; we validated internally that this dilution is performing equally to the manufacturer’s recommended dilution of 1:20; BioLegend), and various combinations of the Abs listed in Supplemental Table I for 20 min at room temperature. After two washes in staining buffer (PBS containing 0.5% FBS and 2 mM EDTA [pH 8.0]), cells were either used for intracellular staining or resuspended into 100 μl of staining buffer and stored at 4°C protected from light for up to 4 h until flow cytometry acquisition.

For intracellular staining, following surface staining cells were fixed for 15 min in paraformaldehyde 4% at room temperature, and washed with freshly prepared permeabilization buffer 1× (from 10× stock; eBioscience). Cells were then stained for intracellular Abs (Supplemental Table I for Ab details) in permeabilization buffer 1× for 30 min at room temperature, washed in staining buffer, resuspended into 100 μl of staining buffer and stored at 4°C protected from light for up to 4 h until flow cytometry acquisition.

For Eomes staining, cells were stained with surface Abs as previously described. After two washes in PBS, cells were fixed and permeabilized with the True-Transcription Factor Buffer Set (BioLegend), according to the manufacturer’s instructions. Subsequently, cells were stained with anti-human Eomes for 30 min at room temperature (Supplemental Table I for Ab details). After two washes in PBS, cells were resuspended in 100 μl of staining buffer and stored at 4°C protected until acquisition.

For tetramer staining, cells were incubated for 1h at 37°C with 30 nM of PE-conjugated tetramer, washed twice with PBS, and then stained with surface Abs as described above. Tetramer is CFP1052–66-DRB5*01:01, an MHC class II tetramer specific for an epitope contained within the CFP10 protein of M. tuberculosis (CFP105266; QAAVVRFQEAANKQK) (24, 25), and was obtained from ImmunAware (Copenhagen, Denmark).

Acquisition was performed on a BD LSR-II Cell Analyzer (BD Biosciences) or on a BD FACS Symphony Cell Sorter (BD Biosciences). Compensation was realized with single-stained beads (UltraComp eBeads, eBioscience) in PBS using the same Ab dilution as for the cell staining. Each Ab was individually titrated for optimum staining. Performance of instruments were checked daily by the Flow Cytometry Core at La Jolla Institute for Immunology with the use of cytometer setup and tracking beads (BD Biosciences), and PMT voltages were manually adjusted for optimum fluorescence detection on each time it was used. A minimum of 150,000 events in the lymphocyte population were recorded for each sample, and data were analyzed using FlowJo software version 10.7.1.

After PBMC thawing, 10 × 106 cells were stained with fixable viability dye eFluor 506 (eBioscience) and with anti-human CD19, CD3, CD4, CD8, CD45RA, and CCR7 (Supplemental Table I for Ab details), as described in the flow cytometry section above. Cell sorting was performed on a BD FACSAria III Cell Sorter (Becton Dickinson). A total of 100,000 nonnaive CD4 T cells (Supplemental Fig. 1A for gating strategy) was sorted into TRIzol LS reagent (Invitrogen) and used for bulk RNA sequencing (RNA-seq). The purity level of the sorted nonnaive CD4 T cells was >99% (Supplemental Fig. 1C).

CTV staining

Cryopreserved PBMC from healthy IGRA+ individuals were quickly thawed, and directly used for CTV staining. A total of 10 μM of working stock was prepared from CellTrace Violet Cell Proliferation Kit (Invitrogen) according to the manufacturer’s protocol. Ten million PBMC were transferred to a 1.5-ml Eppendorf tube. Cells were washed twice with 1× PBS and centrifuged at 2500 rpm for 5 min at room temperature. Cells were resuspended in 1 ml of 1× PBS, and 2 μl of CTV working solution was added. Cells were gently mixed and vortexed before incubating in the dark for 10–12 min at 37°C (with occasional mixing). Cells were washed twice with 1 ml of 20% FBS in 1× PBS for quenching, and resuspended at 10 × 106 cells per milliliter in culture media (RPMI 1640 with L-glutamin and 25 mM HEPES [Omega Scientific], supplemented with 5% human AB serum [GemCell], 1% penicillin streptomycin [Life Technologies], and 1% Glutamax [Life Technologies]).

Culture

Following CTV staining, cells were plated at 0.5 × 106 cells per well in a final volume of 250 µl of media per well (corresponding to 2 × 106 cells per milliliter final cell concentration) in a 96-, U-bottom well plate. We have previously shown that this cell concentration is suitable for in vitro expansion of Ag-specific T cells (26, 27). Cells were stimulated with M. tuberculosis–specific MTB300 peptide pool (28) at 2 μg/ml (final concentration) or an equivalent DMSO concentration (0.3% final). A positive control using plate bound anti-human CD3 Ab (clone OKT3, Invitrogen) and soluble anti-human CD28 Ab (clone CD28.2, BD Biosciences) at 1 ug/ml (final concentration) was also included for each experiment. Cells were incubated at 37°C for up to 14 d. From each well, 125 µl of the culture supernatant (corresponding to half of the total culture volume) was replaced with fresh media every 3–4 d and was supplemented with 0.02 U/μl of IL-2 (Prospec) on days 4, 8, and 12.

Flow cytometry staining

Daily, up to 0.5 × 106 cells were stained with anti-human CD19, CD14, CD3, CD4, CD8, CD45RA, and HLA-DR (Supplemental Table I for Ab details), as described in the flow cytometry section above. Cells were then washed twice with PBS and resuspended in 100 μl of MACS buffer. The 7-AAD viability staining solution (BioLegend) at 1.5 μg/ml was added 30 min before acquisition, as per the manufacturer’s recommendation. No washing was performed after this step to maintain a positive extracellular concentration of 7-AAD, which is necessary to retain the dye intracellularly in nonviable cells.

Gating and sorting of CTV-stained cells

CD4 T cells were defined as 7AADCD14CD19CD8CD3+CD4+ (see gating strategy in Supplemental Fig. 1B). The division gating within CD4 T cells was then defined based on CD4/CTV costaining plots. Because the dye is expected to equally divide between two daughter cells, an approximate 50% loss in fluorescence compared with Div0 was considered as a first division. The Div0 gate was set based on the CTV fluorescence in the baseline prestimulation sample to include all cells; the Div1 lower boundary gate (and Div2+ upper boundary gate) was set at half of the fluorescence of the lower boundary of the Div0 gate (Fig. 5B). At day 8 poststimulation, up to 100,000 of Div0 and Div2+ nonnaive CD45RA CD4 T cells (see gating strategy in Supplemental Fig. 1B) were sorted using a BD Aria II or BD Aria III Cell Sorter (Becton Dickinson) into TRIzol LS reagent (Invitrogen) and used for bulk RNA-seq.

RNA-seq was performed as described previously (22). Briefly, total RNA was purified using an miRNeasy Micro Kit (QIAGEN) and quantified by quantitative PCR, as described previously (29). Purified total RNA (1–5 ng) was amplified following the Smart-Seq2 protocol (16 cycles of cDNA amplification) (30). cDNA was purified using AMPure XP beads (Beckman Coulter). From this step, 1 ng of cDNA was used to prepare a standard Nextera XT Sequencing Library (Nextera XT DNA Sample Preparation Kit and Index Kit; Illumina). Whole-transcriptome amplification and sequencing library preparations were performed in a 96-well format to reduce assay-to-assay variability. Quality-control steps were included to determine total RNA quality and quantity, the optimal number of PCR preamplification cycles, and fragment size selection. Samples that failed quality control were eliminated from further downstream steps. Barcoded Illumina sequencing libraries (Nextera; Illumina) were generated using the automated platform (Biomek FXp). Libraries were sequenced on an HiSeq 2500 Illumina platform to obtain 50-bp single-end reads (TruSeq Rapid kit; Illumina).

RNA-Seq analysis was performed as previously described (22). Briefly, the single-end reads that passed Illumina filters were filtered for reads aligning to transfer RNA, ribosomal RNA, adapter sequences, and spike-in controls. The reads were then aligned to University of California Santa Cruz hg19 reference genome using TopHat (version 1.4.1) (31). DUST scores were calculated with PRINSEQ Lite (v 0.20.3) (32), and low-complexity reads (DUST > 4) were removed from the BAM files. The alignment results were parsed via SAMtools (33) to generate SAM files. Read counts for each genomic feature were obtained with the htseq-count program (version 0.6.0) (34) using the “union” option. After removing absent features (zero counts in all samples), the raw counts were imported to R/Bioconductor package DESeq2 (35) to identify differentially expressed genes among samples. The sequencing data presented in this study were submitted to the Gene Expression Omnibus under accession numbers GSE161829 and GSE162725 (https://www.ncbi.nlm.nih.gov/geo).

Differential expression analysis was done using R Studio version 1.2.5019 and R/Bioconductor package DESeq2 (35). Genes with a transcript per million (TPM) count of zero in >80% samples and genes with a TPM mean value <1 across all samples were filtered out. For nonnaive CD4 T cells isolated ex vivo from ATB, IGRA+ and IGRA individuals, differentially expressed genes were selected based on a p-adjusted value <0.05. For CTV-stained CD4 T cells isolated with the memory T cell proliferation upon Ag exposure (MTP-AE) assay, differentially expressed genes were selected based on a p-adjusted value <0.05 and an absolute log2 fold change >1. Heatmaps were created using R or the software Qlucore using raw counts transformed with the versus function in R. Enrichment for gene ontology (GO) terms associated with biological processes was perform using the online server Enrichr (36, 37). Principal Component Analysis was performed with R, using z-score normalized expression values from the total 35,911 genes detected across all four sorted CD4 T cell populations.

Inference of HLA-DRB5*5 positivity for each participant in the ATB and IGRA+ cohort was done through a personalized HLA allele specific expression pipeline. Each donor’s specific HLA alleles, including HLA-DRB5, was determined using bulk RNA-Seq data with both Optitype version 1.3.2 and PHLAT version 1.1. Once consensus HLA alleles were obtained from both tools, expression quantification was performed using Salmon version 0.11.2. Briefly, the FASTA sequences for each donor’s corresponding computational HLA typing was downloaded from the IMGT-HLA database (38). These FASTA sequences were then used to create a custom Salmon index used for expression quantification. TPM values for each HLA allele were obtained using the Salmon quant method with the following parameters: –rangeFactorizationBins 4, –validateMappings, and –minScoreFraction 1. These parameters were selected to increase quantification accuracy through both alignment validation and restricting alignments to perfect matches. A donor was considered positive for HLA-DRB5*5 expression if expression of the allele was > 1 TPM.

Statistical analyses were performed using GraphPad Prism Software, version 9. Paired datasets were compared using the nonparametric Wilcoxon test, whereas unpaired datasets were compared using the nonparametric Mann–Whitney U test. Any p values <0.05 were considered significant and two-tailed analyses were performed. A p value of overlap between the two gene signatures in (Fig. 1A was defined based on the hypergeometric distribution test (considering the 42,852 transcripts detected in ex vivo nonnaive-sorted CD4 T cells as the total number of genes).

FIGURE 1.

Transcriptomic profiling of circulating nonnaive CD4 T cells in ATB. Nonnaive CD4 T cells were sorted from ATB (n = 24 at diagnosis, and n = 22, 2 mo post-TB treatment), IGRA+ (n = 40) and IGRA (n = 20) individuals as depicted in Supplemental Fig. 1A and their transcriptomic profile defined by bulk RNA-seq. (A) Venn diagram showing overlap between differentially expressed genes in ATB versus IGRA+ and ATB versus IGRA comparisons. (B) Heatmap of the 306 commonly differentially expressed in ATB versus IGRA+ and ATB versus IGRA comparisons. (C) Top 10 GO biological processes terms enriched in the 178 genes upregulated in ATB versus IGRA+ and ATB versus IGRA comparisons. HLA-DRB1 gene expression in the ATB cohort at diagnosis compared with (D) IGRA+ and IGRA cohorts and (E) 2 mo post-TB treatment. * p < 0.05, ** p < 0.01, *** p < 0.001, nonparametric unpaired Mann–Whitney U test (D) and nonparametric paired Wilcoxon test (E).

FIGURE 1.

Transcriptomic profiling of circulating nonnaive CD4 T cells in ATB. Nonnaive CD4 T cells were sorted from ATB (n = 24 at diagnosis, and n = 22, 2 mo post-TB treatment), IGRA+ (n = 40) and IGRA (n = 20) individuals as depicted in Supplemental Fig. 1A and their transcriptomic profile defined by bulk RNA-seq. (A) Venn diagram showing overlap between differentially expressed genes in ATB versus IGRA+ and ATB versus IGRA comparisons. (B) Heatmap of the 306 commonly differentially expressed in ATB versus IGRA+ and ATB versus IGRA comparisons. (C) Top 10 GO biological processes terms enriched in the 178 genes upregulated in ATB versus IGRA+ and ATB versus IGRA comparisons. HLA-DRB1 gene expression in the ATB cohort at diagnosis compared with (D) IGRA+ and IGRA cohorts and (E) 2 mo post-TB treatment. * p < 0.05, ** p < 0.01, *** p < 0.001, nonparametric unpaired Mann–Whitney U test (D) and nonparametric paired Wilcoxon test (E).

Close modal

We have previously shown that it is possible to identify transcriptomic signatures in circulating CD4 T cells that can distinguish healthy IGRA+ from IGRA individuals, reflecting enrichment for M. tuberculosis–specific CD4 T cells in the former cohort (22). To define the immune signature (IMS) of effector CD4 T cells in the context of ATB, we compared the transcriptomic profile of circulating nonnaive CD4 T cells across individuals with ATB (n = 24), previous exposure with M. tuberculosis but no clinical symptoms of ATB (IGRA+, n = 40) and no evidence of M. tuberculosis infection (IGRA, n = 20). We specifically excluded classically defined naive cells that were CCR7+CD45RA+ (see Supplemental Fig. 1A for gating strategy). Effector T cells are expected to have no expression of CCR7 and facultative expression of CD45RA (1012), and will thus fall within this nonnaive compartment. Circulating nonnaive CD4 T cells showed a distinct transcriptomic profile in ATB patients, with 581 and 778 genes differentially expressed compared with IGRA and IGRA+ individuals, respectively (adjusted p < 0.05, Fig. 1A). There was a significant overlap between these gene sets, with 306 genes differentially expressed in both (p value of overlap = 2.2 × 10−16), and we considered this gene list as the expression signature of nonnaive CD4 T cells in ATB (ATB CD4 IMS, Fig. 1A), which we expect to include genes corresponding to recently proliferated CD4 effector T cells. Of the 306 genes from this ATB CD4 IMS, 178 were upregulated in ATB compared with both healthy cohorts (IGRA+ and IGRA), whereas the remaining 128 were downregulated (Fig. 1B). We examined the gene sets for shared functional assignments using the GO biological process function from the online platform Enrichr. The upregulated genes were significantly enriched for 66 GO terms (adjusted p < 0.05). The top 10 included GO terms associated with Ag-specific T cell activation (Ag processing and presentation via MHC class I, TCR signaling pathway) and inflammation (NIK/NfkB signaling, IL-1–mediated signaling pathway) (Fig. 1C). There was no significant enrichment for GO terms in the downregulated genes. HLA-DRB1 was the only previously reported activation marker for T cells among the ATB CD4 IMS, with upregulated gene expression in CD4 T cells of ATB individuals compared with both healthy cohorts (IGRA+ and IGRA) (Fig. 1D). Moreover, in a set of 22 matched paired samples of ATB patients collected at diagnosis and 2–3 mo after the start of anti-TB therapy, HLA-DRB1 expression in CD4 T cells was significantly reduced upon initiation of treatment (Fig. 1E). Thus, RNA expression of HLA-DRB1 on nonnaive CD4 T cells is increased in patients with ATB at diagnosis, before the start of anti-TB therapy.

We performed flow cytometry experiments to test if the observed upregulation of HLA-DR expression in CD4 T cells of ATB patients at the RNA level was also reflected in increased protein expression on the surface of CD4 T cells. We found that the frequency of HLA-DR+ cells was indeed higher in circulating CD4 T cells of individuals with ATB, compared with both healthy cohorts (IGRA+ and IGRA) (Fig. 2A, 2B). To determine whether other T cell activation markers might also be upregulated in circulating CD4 T cells of ATB patients, we explored expression of nine other T cell activation markers, along with HLA-DR, in a smaller set of samples. These markers were selected based on our previous studies aiming at phenotyping Ag-specific T cells in the context of M. tuberculosis (22, 39) but also other viral and bacterial infections (40, 41). We found that three of the selected markers (CD25, OX40, and PDL1) showed increased positive frequency among CD4 T cells in the ATB cohort compared with IGRA individuals (Fig. 2C and Supplemental Fig. 2A). The majority of HLA-DR+ CD4 T cells did not express any of these three markers, and among dual expressers, the highest coexpression was observed for CD25 (Fig. 2D and Supplemental Fig. 2B). The frequency of CD4 T cells with positive expression for each of the remaining six markers (CD38, CD69, CD137, CD153, CD154, and PD1) did not show any change across M. tuberculosis–infected and uninfected cohorts (Fig. 2E). Thus, the frequency of HLA-DR+ CD4 T cells is increased in the peripheral blood of ATB patients but not healthy IGRA+ or IGRA individuals.

FIGURE 2.

Increased frequency of HLA-DR+ single-expressers in circulating CD4 T cells of ATB patients. (A) Representative flow cytometry staining plot of HLA-DR expression in CD4 T cells of ATB, IGRA+ and IGRA individuals. Frequency of (B) HLA-DR+ cells; (C) CD25+, OX40+, and PDL1+ cells; (D) HLA-DR+ cells coexpressing CD25, OX40, or PDL1; (E) CD38+, CD69+, CD137+, CD153+, CD154+, and PD1+ cells in CD4 T cells of ATB, IGRA+ and IGRA individuals determined by flow cytometry. (A and B) combined data from n = 15 ATB, n = 11 IGRA+, and n = 11 IGRA individuals. (C–E) combined data from n = 4 ATB, n = 5 IGRA+, and n = 5 IGRA individuals. CD4 T cells were gated as CD3+CD4+CD8CD19CD14 in the live singlet gate of PBMC. * p < 0.05, ** p < 0.01, nonparametric unpaired Mann–Whitney U test.

FIGURE 2.

Increased frequency of HLA-DR+ single-expressers in circulating CD4 T cells of ATB patients. (A) Representative flow cytometry staining plot of HLA-DR expression in CD4 T cells of ATB, IGRA+ and IGRA individuals. Frequency of (B) HLA-DR+ cells; (C) CD25+, OX40+, and PDL1+ cells; (D) HLA-DR+ cells coexpressing CD25, OX40, or PDL1; (E) CD38+, CD69+, CD137+, CD153+, CD154+, and PD1+ cells in CD4 T cells of ATB, IGRA+ and IGRA individuals determined by flow cytometry. (A and B) combined data from n = 15 ATB, n = 11 IGRA+, and n = 11 IGRA individuals. (C–E) combined data from n = 4 ATB, n = 5 IGRA+, and n = 5 IGRA individuals. CD4 T cells were gated as CD3+CD4+CD8CD19CD14 in the live singlet gate of PBMC. * p < 0.05, ** p < 0.01, nonparametric unpaired Mann–Whitney U test.

Close modal

Effector T cells originate from Ag-specific T cells that start to divide upon TCR activation. To investigate if the increased population of circulating HLA-DR+ CD4 T cells in ATB recognize M. tuberculosis Ags, we stained cells with an MHC class II (DRB5*01:01) tetramer loaded with a CFP10 epitope (24, 25). Tetramer-positive (tet+) M. tuberculosis–specific CD4 T cells could be identified in five out of six IGRA+ individuals and four out of five ATB individuals tested that expressed DRB5*01:01 (Fig. 3A, frequency above the limit of detection of 0.01% of total CD4 T cells). Average frequencies were higher in the ATB cohort compared with IGRA+ (Fig. 3A, 0.06% and 0.02% tet+ among total CD4 T cells, in ATB versus IGRA+, respectively). In each participant, we looked for costaining of the tetramer with HLA-DR. We also included the three activation markers that also showed significant upregulation in circulating CD4 T cells in ATB patients compared with IGRA individuals, namely CD25, OX40, or PDL1 (Fig. 2C). In ATB patients, the vast majority (58–88%) of tet+ cells expressed high levels of HLA-DR (Fig. 3B, 3C). Less than 20% of tet+ cells expressed CD25 or PDL1, and no tet+ cells expressed OX40 (Fig. 3B, 3C). In contrast, when looking at IGRA+ individuals, less than 3% of tet+ cells expressed HLA-DR (Fig. 3D, 3E). Thus, HLA-DR marks M. tuberculosis–specific CD4 T cells in patients with ATB, but not IGRA+ individuals, consistent with them being recently activated effector cells.

FIGURE 3.

HLA-DR is expressed on Ag-specific CD4 T cells in ATB but not healthy IGRA+ individuals. M. tuberculosis–specific CD4 T cells were identified by flow cytometry using CFP1052–66–DRB5*01:01, an MHC class II tetramer specific for an epitope contained within the CFP10 protein of M. tuberculosis (CFP105266; QAAVVRFQEAANKQK) (24, 25). (A) Frequency of tetramer-positive (tet+) cells in ATB (n = 4) and IGRA+ (n = 5) individuals that were defined as HLA-DRB5*01:01 positive from the cohort analyzed in (Fig. 1. One ATB and one IGRA+ participants had undetectable levels of tet+ cells (less than 0.01% of total CD4 T cells) and were excluded from the analysis. (B) Representative costaining plots and (C) frequency of tet+ cells coexpressing CD25, HLA-DR, OX40, or PDL1 in ATB. (D) Representative costaining plots and (E) frequency of tet+ cells coexpressing HLA-DR in ATB compared with IGRA+ individuals. * p < 0.05, nonparametric unpaired Mann–Whitney U test.

FIGURE 3.

HLA-DR is expressed on Ag-specific CD4 T cells in ATB but not healthy IGRA+ individuals. M. tuberculosis–specific CD4 T cells were identified by flow cytometry using CFP1052–66–DRB5*01:01, an MHC class II tetramer specific for an epitope contained within the CFP10 protein of M. tuberculosis (CFP105266; QAAVVRFQEAANKQK) (24, 25). (A) Frequency of tetramer-positive (tet+) cells in ATB (n = 4) and IGRA+ (n = 5) individuals that were defined as HLA-DRB5*01:01 positive from the cohort analyzed in (Fig. 1. One ATB and one IGRA+ participants had undetectable levels of tet+ cells (less than 0.01% of total CD4 T cells) and were excluded from the analysis. (B) Representative costaining plots and (C) frequency of tet+ cells coexpressing CD25, HLA-DR, OX40, or PDL1 in ATB. (D) Representative costaining plots and (E) frequency of tet+ cells coexpressing HLA-DR in ATB compared with IGRA+ individuals. * p < 0.05, nonparametric unpaired Mann–Whitney U test.

Close modal

To further characterize HLA-DR+ CD4 effector T cells in the ATB cohort, we examined their phenotype using flow cytometry. Effector T cells are expected to downregulate CCR7 (1012). As expected, we found that the CD45RACCR7 phenotype was significantly overrepresented in HLA-DR+ compared with HLA-DR CD4 T cells. CD45RACCR7 cells encompassed 51–74% of HLA-DR+ cells but only 7–21% of HLA-DR CD4 T cells (Figs. 4A, 4B). Key functions of effector T cells are direct killing of infected cells or indirect modulation of the immune response through the expression of cytokines. We compared the expression of cytotoxic molecules, as well as spontaneous cytokine production, in HLA-DR+ versus HLA-DR CD4 T cells from ATB patients. We found that HLA-DR+ CD4 T cells were enriched for cells expressing the cytolytic granule granzyme B (GzmB) (19–72% versus 2–51% GzmB+ cells in HLA-DR+ versus HLA-DR), the degranulation marker CD107a (1–19% versus 0.3–6% CD107a+ cells in HLA-DR+ versus HLA-DR), the activation and exhaustion marker Tim3 (0.5–4.9% versus 0.1–0.8% Tim3+ cells in HLA-DR+ versus HLA-DR), and the cytotoxic transcription factor Eomes (13–34% versus 1–7% Eomes+ cells in HLA-DR+ versus HLA-DR), in comparison with HLA-DR CD4 T cells (Fig. 4C). HLA-DR+ CD4 T cells also had a higher proportion of cells spontaneously producing TNF-α (0.1–2.1% versus 0.1–0.6% TNF-α+ cells in HLA-DR+ versus HLA-DR), but not IFN-γ or IL-17 in comparison with HLA-DR CD4 T cells (Fig. 4D). To address whether these phenotypic features were specific to ATB or a general feature of HLA-DR+ CD4 T cells, we compared the differentiation and effector phenotype of HLA-DR+ CD4 T cells in ATB, IGRA+, and IGRA individuals. The differentiation phenotype (assessed by CD45RA and CCR7 coexpression) was not statistically different between HLA-DR+ CD4 T cells of ATB and IGRA+ individuals, with a predominant CD45RACCR7 phenotype (p > 0.05, Supplemental Fig. 3A). A higher proportion of HLA-DR+ CD4 T cells expressed GzmB and Tim3 in ATB compared with IGRA but not IGRA+ individuals (Supplemental Fig. 3B). Similar proportion of cells expressing CD107a, Eomes, or any of the three measured cytokines in HLA-DR+ CD4 T cells was observed across all three cohorts (Supplemental Fig. 3B and Supplemental Fig. 3C). Thus, in ATB patients, circulating HLA-DR+ CD4 T cells present characteristics of effector T cells with expression of markers associated with effector phenotype, cytotoxic function, and recent activation.

FIGURE 4.

HLA-DR expression marks a subset of CD4 T cells with an effector phenotype. (A) Representative flow cytometry staining plot of CD45RA and CCR7 expression in HLA-DR+ versus HLA-DR CD4 T cells in ATB. (B) Frequency of CD45RA/CCR7 boolean phenotypes in HLA-DR+ versus HLA-DR CD4 T cells in ATB patients (n = 5) determined by flow cytometry. Frequency of positive cells for (C) cytotoxic and activation markers GzmB, CD107a, Tim3, Eomes, and (D) spontaneous production of the cytokines TNF-α, IFN-γ, and IL-17 in HLA-DR+ versus HLA-DR CD4 T cells of ATB patients (n = 5 for Eomes, n = 10 for all others), determined by flow cytometry. * p < 0.05, ** p < 0.01, nonparametric paired Wilcoxon test.

FIGURE 4.

HLA-DR expression marks a subset of CD4 T cells with an effector phenotype. (A) Representative flow cytometry staining plot of CD45RA and CCR7 expression in HLA-DR+ versus HLA-DR CD4 T cells in ATB. (B) Frequency of CD45RA/CCR7 boolean phenotypes in HLA-DR+ versus HLA-DR CD4 T cells in ATB patients (n = 5) determined by flow cytometry. Frequency of positive cells for (C) cytotoxic and activation markers GzmB, CD107a, Tim3, Eomes, and (D) spontaneous production of the cytokines TNF-α, IFN-γ, and IL-17 in HLA-DR+ versus HLA-DR CD4 T cells of ATB patients (n = 5 for Eomes, n = 10 for all others), determined by flow cytometry. * p < 0.05, ** p < 0.01, nonparametric paired Wilcoxon test.

Close modal

To further test if HLA-DR+ CD4 T cells in ATB patients are indeed effector T cells, we wanted to examine if this phenotype is consistent with recent proliferation after Ag-specific activation, which is a fundamental hallmark of effector T cells. Identifying recently proliferated cells in humans in vivo is challenging. One commonly used approach is detecting the presence of the protein Ki67, which identifies proliferating T cells in human blood samples, but it can only detect cells that are actively undergoing division, or shortly thereafter, and not the ones that have recently completed cell division (42). Detection of Ki67 also requires nuclear staining and thus cell fixation, which limits downstream analyses that can be performed on positive cells. To address this, we designed an in vitro assay aiming at mimicking MTP-AE. In the MTP-AE assay, PBMC samples from healthy IGRA+ individuals (which are expected to contain a significant number of M. tuberculosis–specific memory CD4 T cells) were stained using the proliferation dye CTV. CD4 T cells were then stimulated with a pool of M. tuberculosis–derived peptide epitopes (28), simulating Ag exposure. IL-2 was given on days 4, 8, and 12 as an additional signal for proliferation. Cells were cultured in vitro for 14 d, and the level of CTV staining was monitored daily to identify divided cells. The MTP-AE assay mimics how memory M. tuberculosis–specific CD4 T cells are activated, and start proliferating into differentiated effector cells, similar to what we would expect in vivo in the development of an active infection.

We applied the MTP-AE assay on PBMC from 5 IGRA+ individuals. CD4 T cells were partitioned into those that had not divided (Div0), those that had divided once (Div1 group) and those that had divided at least twice (Div2+ group) based on CTV fluorescence (Fig. 5A, 5B). Along with CTV, we also monitored the surface expression of HLA-DR in CD4 T cells using flow cytometry to divide them into HLA-DR+ and HLA-DR CD4 T cells. A significant number of proliferating cells could be detected from day 4 onwards, with Div1 and Div2+ peaking at day 7 and day 11 poststimulation, respectively (Fig. 5C). Bivariate flow cytometry plots of HLA-DR with CTV and a side-by-side comparison of the frequency of HLA-DR+ cells within undivided (Div0) versus divided (Div1 and Div2+) CD4 T cells showed HLA-DR expression was mostly restricted to divided CD4 T cells (Fig. 5D, 5E).

FIGURE 5.

HLA-DR expression marks recently divided CD4 T cells upon M. tuberculosis Ag exposure. Using our in vitro model of M. tuberculosis–specific T cell proliferation assay, PBMC of healthy IGRA+ individuals were stained with proliferation dye CTV and in vitro stimulated with the MTB300 peptide pool (28) for 14 d. Each day post stimulation, CD4 T cells were identified as either undivided (Div0), divided once (Div1), or divided more than once (Div2+) by flow cytometry. (A) Representative costaining plots of CD4 and CTV expression before (day 0), 4 d, and 10 d post–MTB300 stimulation within CD4 T cells. (B) Representative histogram of CTV expression at 4 d and 10 d post–MTB300 stimulation. (C) Number of CD4 T cells within each CTV division category in function of days post MTB300 stimulation. (D) Representative costaining plots of HLA-DR and CTV expression before (day 0), 4 d, and 10 d post–MTB300 stimulation within CD4 T cells. (E) Frequency of HLA-DR+ undivided (Div0) and HLA-DR+ divided (Div1/Div2+) CD4 T cells in function of days post–MTB300 stimulation. (A–E) Combined data from n = 5 IGRA+ individuals. (F) Principal component analysis based on all genes expressed in sorted CD4 T cell populations and (G) heatmap representing the expression of the top 25 genes upregulated in Div2+ versus Div0 nonnaive CD4 T cells upon MTB300 stimulation. Genes highlighted in blue have a flow cytometric Ab available for measurement of their protein expression. Cells were sorted from 5 IGRA+ individuals at day 8 poststimulation with either DMSO or MTB300 (see gating strategy in Supplemental Fig. 1B), and their transcriptomic profile determined by RNA-seq. (H) Fold change in positive frequency of selected markers to identify recently divided cells [blue genes in (G), excluding GZMB] in HLA-DR+ compared with HLA-DR CD4 T cells in ATB patients (n = 10 for CCL3, CTLA4, TNFRSF18, CD82, OX40, and CCL4; n = 4 for CD38 and CD25) determined by flow cytometry. Significant markers (p < 0.05) are represented in green. (I) Frequency of positive cells for CCL3, CD38, and OX40 in HLA-DR+ CD4 T cells of ATB [similar cohort to (H)], IGRA+ (n = 6) and IGRA (n = 6) individuals. *p < 0.05, **p < 0.01, *** p < 0.001, **** p < 0.0001, nonparametric paired Wilcoxon test.

FIGURE 5.

HLA-DR expression marks recently divided CD4 T cells upon M. tuberculosis Ag exposure. Using our in vitro model of M. tuberculosis–specific T cell proliferation assay, PBMC of healthy IGRA+ individuals were stained with proliferation dye CTV and in vitro stimulated with the MTB300 peptide pool (28) for 14 d. Each day post stimulation, CD4 T cells were identified as either undivided (Div0), divided once (Div1), or divided more than once (Div2+) by flow cytometry. (A) Representative costaining plots of CD4 and CTV expression before (day 0), 4 d, and 10 d post–MTB300 stimulation within CD4 T cells. (B) Representative histogram of CTV expression at 4 d and 10 d post–MTB300 stimulation. (C) Number of CD4 T cells within each CTV division category in function of days post MTB300 stimulation. (D) Representative costaining plots of HLA-DR and CTV expression before (day 0), 4 d, and 10 d post–MTB300 stimulation within CD4 T cells. (E) Frequency of HLA-DR+ undivided (Div0) and HLA-DR+ divided (Div1/Div2+) CD4 T cells in function of days post–MTB300 stimulation. (A–E) Combined data from n = 5 IGRA+ individuals. (F) Principal component analysis based on all genes expressed in sorted CD4 T cell populations and (G) heatmap representing the expression of the top 25 genes upregulated in Div2+ versus Div0 nonnaive CD4 T cells upon MTB300 stimulation. Genes highlighted in blue have a flow cytometric Ab available for measurement of their protein expression. Cells were sorted from 5 IGRA+ individuals at day 8 poststimulation with either DMSO or MTB300 (see gating strategy in Supplemental Fig. 1B), and their transcriptomic profile determined by RNA-seq. (H) Fold change in positive frequency of selected markers to identify recently divided cells [blue genes in (G), excluding GZMB] in HLA-DR+ compared with HLA-DR CD4 T cells in ATB patients (n = 10 for CCL3, CTLA4, TNFRSF18, CD82, OX40, and CCL4; n = 4 for CD38 and CD25) determined by flow cytometry. Significant markers (p < 0.05) are represented in green. (I) Frequency of positive cells for CCL3, CD38, and OX40 in HLA-DR+ CD4 T cells of ATB [similar cohort to (H)], IGRA+ (n = 6) and IGRA (n = 6) individuals. *p < 0.05, **p < 0.01, *** p < 0.001, **** p < 0.0001, nonparametric paired Wilcoxon test.

Close modal

To confirm that the phenotype of HLA-DR+ CD4 T cells reflects that of recently divided cells, we investigated the protein expression of several markers for recently divided cells in HLA-DR+ versus HLA-DR CD4 T cells. To define the optimal markers that can identify recently divided cells, we sorted Div0 and Div2+ nonnaive CD45RA CD4 T cells at 8 d poststimulation and analyzed their transcriptomic profile by RNA-seq (Supplemental Fig. 1B for gating strategy). We selected day 8 for sorting as it was the timepoint with the highest number of Div2+ cells before the sharp drop in undivided cells that occurred at day 9 poststimulation, likely because of cell death. Unsupervised analysis showed HLA-DR+ and Div2+ CD4 T cells had a similar gene expression profile in comparison with HLA-DR and Div0 CD4 T cells (Fig. 5F). To translate these transcriptomic similarities at the protein level, we selected the most abundant 25 genes that were upregulated in Div2+ cells compared with Div0 cells (Fig. 5G, adjusted p <0.05 and log fold change >2, ranked by decreasing TPM value). Among these 25 genes was GzmB, which was previously found to be upregulated at the protein level in HLA-DR+ compared with HLA-DR CD4 T cells (Fig. 4C). An additional eight genes for which a fluorochrome-conjugated Ab was commercially available were selected to measure their protein expression in HLA-DR+ and HLA-DR CD4 T cells of ATB patients by flow cytometry. For each marker, we calculated a fold change expression between the frequency of positive cells within HLA-DR+ and HLA-DR CD4 T cells. Out of the eight measured markers, only OX40 and CCL4 did not show positive frequency differences between the HLA-DR+ and HLA-DR populations (Fig. 5H). CCL3, CD38, CTLA4, CD25, TNFRSF18, and CD82 all showed significant higher frequency of positive cells in HLA-DR+ compared with HLA-DR CD4 T cells (Fig. 5H, Supplemental Fig. 3E). The highest fold change between HLA-DR+ and HLA-DR populations was observed for CCL3, followed by CD38, CTLA4, and CD25 (Fig. 5H). These four markers also showed very low frequency of positive cells within HLA-DR cells compared with HLA-DR+, indicating a high specificity for the HLA-DR+ subset (Supplemental Fig. 3E). The frequency of cells expressing CCL3, CD38, and OX40 was also significantly increased in HLA-DR+ cells of ATB patients compared with the IGRA+ cohort (as well as IGRA individuals for CCL3 and CD38) (Fig. 5I). The frequency of positive cells within HLA-DR+ CD4 T cells for the other measured markers was unchanged across all three cohorts (Supplemental Fig. 3D). Taken together, these results indicate that HLA-DR expression is a hallmark for CD4 T cells that have recently divided after exposure to M. tuberculosis Ags.

In this study, we set out to determine the phenotype of effector CD4 T cells that are generated during ATB disease as a result of Ag encounter. Effector CD4 T cells are expected to form a predominantly short-lived, but effective, arm of the immune response to fight active infections. Through multiple lines of evidence, we found that HLA-DR expression was a marker of such effector cells, and that HLA-DR+ CD4 T cells were increased in individuals with ATB.

This is not the first association between HLA-DR and CD4 T cells in the context of TB. In bacillus Calmette–Guérin–vaccinated infants and IGRA+ adolescents, frequencies of activated HLA-DR+ CD4 T cells were associated with increased active disease risk (43) and decreased at 27 wk postvaccination (44). In terms of Ag-specific CD4 T cells, several reports have shown that HLA-DR expression on cytokine producing CD4 T cells after M. tuberculosis–specific in vitro stimulation can distinguish ATB from healthy IGRA+ individuals (4549) and can also be predictive of progression to ATB (C. A. M. Mpande, M. Musvosvi, V. Rozot, B. Mosito, T. D. Reid, C. Schreuder, T. Lloyd, N. Bilek, H. Huang, G. Obermoser, et al., manuscript posted on medRxiv, DOI: 10.1101/2020.06.26.20135665) and sputum negative conversion upon TB treatment (50, 51). Thus, our data corroborates these findings and establishes that the frequency of HLA-DR+ cells is also increased in M. tuberculosis–specific CD4 T cells isolated directly ex vivo in the context of ATB. More importantly, we demonstrated for the first time that HLA-DR marks a population of CD4 T cells with characteristics of effector cells, providing a seminal reason for why clinicians and immunologists should investigate more closely the HLA-DR+ CD4 T cell population in M. tuberculosis–infected individuals.

We found that HLA-DR+ CD4 T cells presented a cytotoxic phenotype with higher proportion of cells expressing the cytotoxic markers GzmB, Eomes, Tim3, CD107a. This is not the first association between M. tuberculosis–specific CD4 T cells and markers of cytotoxicity. CD4 T cells have been shown to upregulate the expression of granzymes, granulysin, and perforin after M. tuberculosis–specific in vitro stimulation and the ability to lyse M. tuberculosis–infected monocytes (52, 53). More recently, M. tuberculosis–specific CD4 T cells isolated from IGRA+ individuals showed increased expression of GzmB and CD107a after Ag-specific in vitro stimulation and expansion (54). Altogether, our and previous observations suggest cytotoxicity is a phenotypic characteristic of M. tuberculosis–specific effector T cells, and that this function might be important in driving protective immunity.

In terms of cytokines, we showed that HLA-DR+ CD4 T cells had a higher proportion of cells expressing the cytokine TNF-α, but not IFN-γ or IL-17. The frequency of TNF-α single-positive M. tuberculosis–specific CD4 T cells has been positively associated with ATB and suggested as a diagnostic marker to distinguish ATB from healthy IGRA+ individuals (55). TNF-α has a critical role in cell death and cytotoxicity (56), corroborating our findings on the association between M. tuberculosis–specific effector CD4 T cells and a cytotoxic phenotype. The absence of IFN-γ is surprising because it is known to be a major driver of CD4 T cell protective immunity in TB. A previous study has shown that cytokine production (including IFN-γ and IL-17) after PMA/ionomycin stimulation was greater in HLA-DR+ compared with HLA-DR CD4 T cells from ATB patients (57). In this study, we only assessed spontaneous cytokine production directly ex vivo. Thus, HLA-DR+ CD4 T cells in our ATB cohort might nonetheless have a greater ability to produce IFN-γ and other cytokines upon in vitro stimulation in comparison with HLA-DR CD4 T cells.

We established an in vitro assay using stimulation with an epitope megapool in an appropriate cytokine environment to mimic Ag encounter, combined with the use of a proliferation dye to identify dividing T cells in response to Ag exposure. We named this assay MTP-AE. Whereas the use of proliferation dyes is not novel, this is, to our knowledge, the first attempt to leverage them to assess the transcriptomic profile of Ag-specific CD4 T cells that have recently divided. We used a proliferation dye in combination with flow cytometry, and RNA-Seq to determine that the phenotype of HLA-DR+ CD4 T cells significantly overlaps with that of recently divided cells. The MTP-AE assay can be straightforwardly modified using a stimulus of distinct nature (e.g., whole organism, single-protein, or peptide, etc.) or distinct antigenic specificity, as long as the starting PBMC population contain the corresponding Ag-specific memory T cell population that can be activated. For instance, it could help identifying similarities and differences in the transcriptomic profile of proliferating M. tuberculosis–specific CD4 T cells after stimulation with whole organisms such as heat-killed M. tuberculosis, in comparison with the M. tuberculosis–specific peptide pool that was used in this study. It could also be used with stimuli specific to other pathogens, for instance CMV or Influenza, in samples from M. tuberculosis–infected and uninfected individuals, to determine if HLA-DR expression in proliferating CD4 T cells can be generalized to other pathogens, or whether it is dependent on the presence of M. tuberculosis infection. Additionally, this assay can be easily applied to genomic analyses other than transcriptomics, for instance to decipher the epigenetic profile of HLA-DR+ CD4 T cells, and how it varies across cohorts and over time in M. tuberculosis–infected individuals. Therefore, this assay is a powerful method to study the molecular profile of recently divided Ag-specific memory T cells, and thus effector T cells. The main caveat of this in vitro expansion assay is the impossibility to distinguish between cells that are activated versus cells that are more prone to survival. An independent validation of the results with phenotyping assays directly ex vivo in patients’ samples can overcome this hurdle. For instance, in this study using the MTP-AE assay we identified that HLA-DR marks M. tuberculosis–specific proliferating CD4 T cells in vitro. But we also found that the frequency of HLA-DR+ CD4 T cells is increased ex vivo in ATB patients, and that this cell population is associated with increased frequency of cells expressing activation and cytotoxic markers, suggesting that in this context, HLA-DR marks activation rather than survival.

Whether HLA-DR+ CD4 T cells reflect circulating CD4 T cells or recirculate after egressing from peripheral tissues remains to be determined. The fact that HLA-DR expression is strongly increased in tissue resident memory T cells present in the BAL and lung of ATB infected patients (58) suggests circulating HLA-DR+ CD4 T cells might originate from M. tuberculosis–infected lung tissues or their draining lymph nodes, and then egress into the peripheral circulation. This hypothesis is consistent with the idea that HLA-DR+ CD4 T cells are effector T cells, and are thus expected to be generated at or near the site of infection, where Ag is present.

Although they were more prevalent in ATB, we also observed HLA-DR+ CD4 T cells in both healthy cohorts (IGRA+ and IGRA), and most phenotypic features of this cell population were shared across all cohorts. Circulating HLA-DR+ CD4 T cells in IGRA+ individuals may also represent M. tuberculosis–specific effector T cells because humans with M. tuberculosis sensitization may not control the bacterium completely such that these T cells are exposed to M. tuberculosis Ags at a low level or from time to time. They might also represent effector T cells that are specific to other pathogens and that have been recently activated. Alternatively, they could mark a noneffector T cell subset that constitutively express HLA-DR and is ubiquitously present in circulating CD4 T cells in humans. Indeed, even in ATB patients, not all HLA-DR+ CD4 T cells showed positive expression of cytotoxic markers or cytokines. Thus, only a subset of HLA-DR–expressing cells might represent effector T cells. Single-cell RNA-seq of HLA-DR+ CD4 T cells will shed light on the heterogeneity of this cell population.

The phenotypic features specific of HLA-DR+ CD4 T cells in ATB compared with IGRA+ or IGRA cohorts included increased frequency of positive cells expressing CCL3 or CD38 (Fig 5I). These two markers also showed the highest differential fold change frequency between HLA-DR+ and HLA-DR CD4 T cells (Fig 5H). Interestingly, the frequency of CD38+ cells in total CD4 T cells was not different across all three cohorts (Fig 2E), suggesting CD38 is useful in discriminating CD4 T cells in ATB only in conjunction with HLA-DR. CD38 was measured along with HLA-DR in several studies aiming at phenotyping CD4 T cells in the context of ATB and was also found to be upregulated in M. tuberculosis–specific CD4 T cells of ATB compared with IGRA+-infected individuals (45, 49, 51). More recently, in patients with acute dengue fever, CCL3 and CD38 were identified as significantly upregulated at the RNA level in Ag-specific CD4 T cells (59), as well as HLA-DR (data not shown). In addition to CCL3 and CD38, we also found increased frequency of positive cells expressing GzmB and Tim3 in HLA-DR+ CD4 T cells of ATB compared with IGRA- (Supplemental Fig. 3B), and increased frequency of positive cells expressing OX40 in HLA-DR+ CD4 T cells of ATB compared with IGRA+ individuals (Fig 5I). Altogether, these additional phenotypic markers might be extremely useful in conjunction with HLA-DR to further delineate effector T cell subsets, and better distinguishing them from effector memory T cells. Because our initial panel was restricted to activation markers that we typically use for our T cell studies (Fig 2 and Refs. 22, 3941), our coexpression findings could also be extended to a broader range of T cell activation markers, such as CD44, which is a potent marker for Ag-experienced T cells (60, 61).

Lastly, the upregulation of HLA-DR in circulating total and Ag-specific T cells has been described in several other human infection models including EBV (62), HIV (63), dengue (64) and vaccination models of yellow fever (12, 65), smallpox (65), and malaria (66). More recently, the frequency of HLA-DR+ cells was shown to be increased in circulating total (6770) and SARS-CoV-2–specific (71, 72) CD4 T cells of acute COVID-19 patients and correlated with disease severity (6870, 72). Thus, HLA-DR might be a useful marker for identifying effector T cells and monitoring immune responses not only in the context of TB, but also many other infection and vaccination models. Overall, the methodological approach and results reported in this study represents a stepping stone to facilitate the investigation of effector T cells in humans.

The gene expression data included in this study were submitted to the Gene Expression Omnibus under accession numbers GSE161829 (Fig. 1) and GSE162725 (Fig. 5) (https://www.ncbi.nlm.nih.gov/geo). All other relevant data supporting the key findings of this study are available within the article and its Supplemental Material files or from the corresponding author upon reasonable request.

We thank the Flow Cytometry Core, the Sequencing Core, and the Bioinformatics Core facilities at La Jolla Institute for Immunology for technical assistance, and Shane Crotty and Carolyn Mobermacher for helpful discussion on the manuscript.

This work was supported by the National Institute of Allergy and Infectious Diseases of the National Institute of Health (Grant U19 AI118626).

R. Tippalagama, M.P., B.P., and J.G.B. conceived and designed the study. A.D.d., S.P., D.V., B.G., N.D.S.G., D.A., T.J.S., R.H.G., M.S., and R. Taplitz provided samples. C.S.L.A., G.S., and P.V. provided technical resources. R. Tippalagama, A. Singhania, P.D., A.C., and J.G.B. conducted the experiments and/or analyzed the data. A. Sette and B.P. provided funding. R. Tippalagama, B.P., and J.G.B. led the data analysis and interpretation with input from all coauthors. R. Tippalagama, B.P., and J.G.B. wrote the manuscript and all authors edited the manuscript.

The sequences presented in this article have been submitted to the Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo) under accession numbers GSE161829 and GSE162725.

The online version of this article contains supplemental material.

Abbreviations used in this article

     
  • ATB

    active TB

  •  
  • CTV

    Cell Trace Violet

  •  
  • GO

    gene ontology

  •  
  • GzmB

    granzyme B

  •  
  • IMS

    immune signature

  •  
  • RNA-seq

    RNA sequencing

  •  
  • TB

    tuberculosis

  •  
  • TPM

    transcript per million

1.
Kaech
S. M.
,
E. J.
Wherry
,
R.
Ahmed
.
2002
.
Effector and memory T-cell differentiation: implications for vaccine development.
Nat. Rev. Immunol.
2
:
251
262
.
2.
Bradley
L. M.
,
D. D.
Duncan
,
K.
Yoshimoto
,
S. L.
Swain
.
1993
.
Memory effectors: a potent, IL-4-secreting helper T cell population that develops in vivo after restimulation with antigen.
J. Immunol.
150
:
3119
3130
.
3.
Cui
W.
,
S. M.
Kaech
.
2010
.
Generation of effector CD8+ T cells and their conversion to memory T cells.
Immunol. Rev.
236
:
151
166
.
4.
Kaech
S. M.
,
W.
Cui
.
2012
.
Transcriptional control of effector and memory CD8+ T cell differentiation.
Nat. Rev. Immunol.
12
:
749
761
.
5.
Strutt
T. M.
,
K. K.
McKinstry
,
N. B.
Marshall
,
A. M.
Vong
,
R. W.
Dutton
,
S. L.
Swain
.
2013
.
Multipronged CD4(+) T-cell effector and memory responses cooperate to provide potent immunity against respiratory virus.
Immunol. Rev.
255
:
149
164
.
6.
Sallusto
F.
,
A.
Lanzavecchia
.
2009
.
Heterogeneity of CD4+ memory T cells: functional modules for tailored immunity.
Eur. J. Immunol.
39
:
2076
2082
.
7.
Cruz-Adalia
A.
,
G.
Ramirez-Santiago
,
C.
Calabia-Linares
,
M.
Torres-Torresano
,
L.
Feo
,
M.
Galán-Díez
,
E.
Fernández-Ruiz
,
E.
Pereiro
,
P.
Guttmann
,
M.
Chiappi
, et al
2014
.
T cells kill bacteria captured by transinfection from dendritic cells and confer protection in mice.
Cell Host Microbe
15
:
611
622
.
8.
Marshall
N. B.
,
S. L.
Swain
.
2011
.
Cytotoxic CD4 T cells in antiviral immunity.
J. Biomed. Biotechnol.
2011
:
954602
.
9.
Swain
S. L.
,
K. K.
McKinstry
,
T. M.
Strutt
.
2012
.
Expanding roles for CD4+ T cells in immunity to viruses.
Nat. Rev. Immunol.
12
:
136
148
.
10.
Mahnke
Y. D.
,
T. M.
Brodie
,
F.
Sallusto
,
M.
Roederer
,
E.
Lugli
.
2013
.
The who’s who of T-cell differentiation: human memory T-cell subsets.
Eur. J. Immunol.
43
:
2797
2809
.
11.
Restifo
N. P.
,
L.
Gattinoni
.
2013
.
Lineage relationship of effector and memory T cells.
Curr. Opin. Immunol.
25
:
556
563
.
12.
Blom
K.
,
M.
Braun
,
M. A.
Ivarsson
,
V. D.
Gonzalez
,
K.
Falconer
,
M.
Moll
,
H. G.
Ljunggren
,
J.
Michaëlsson
,
J. K.
Sandberg
.
2013
.
Temporal dynamics of the primary human T cell response to yellow fever virus 17D as it matures from an effector- to a memory-type response.
J. Immunol.
190
:
2150
2158
.
13.
Okada
R.
,
T.
Kondo
,
F.
Matsuki
,
H.
Takata
,
M.
Takiguchi
.
2008
.
Phenotypic classification of human CD4+ T cell subsets and their differentiation.
Int. Immunol.
20
:
1189
1199
.
14.
Appay
V.
,
R. A.
van Lier
,
F.
Sallusto
,
M.
Roederer
.
2008
.
Phenotype and function of human T lymphocyte subsets: consensus and issues.
Cytometry A
73
:
975
983
.
15.
World Health Organization
.
2019
.
Global tuberculosis report 2019.
.
16.
Vynnycky
E.
,
P. E.
Fine
.
1997
.
The natural history of tuberculosis: the implications of age-dependent risks of disease and the role of reinfection.
Epidemiol. Infect.
119
:
183
201
.
17.
Drain
P. K.
,
K. L.
Bajema
,
D.
Dowdy
,
K.
Dheda
,
K.
Naidoo
,
S. G.
Schumacher
,
S.
Ma
,
E.
Meermeier
,
D. M.
Lewinsohn
,
D. R.
Sherman
.
2018
.
Incipient and subclinical tuberculosis: A clinical review of early stages and progression of infection.
Clin. Microbiol. Rev.
31
:
e00021-18
.
18.
Barnes
P. F.
,
A. B.
Bloch
,
P. T.
Davidson
,
D. E.
Snider
Jr
.
1991
.
Tuberculosis in patients with human immunodeficiency virus infection.
N. Engl. J. Med.
324
:
1644
1650
.
19.
Flynn
J. L.
,
J.
Chan
.
2001
.
Immunology of tuberculosis.
Annu. Rev. Immunol.
19
:
93
129
.
20.
Lindestam Arlehamn
C. S.
,
A.
Gerasimova
,
F.
Mele
,
R.
Henderson
,
J.
Swann
,
J. A.
Greenbaum
,
Y.
Kim
,
J.
Sidney
,
E. A.
James
,
R.
Taplitz
, et al
2013
.
Memory T cells in latent Mycobacterium tuberculosis infection are directed against three antigenic islands and largely contained in a CXCR3+CCR6+ Th1 subset.
PLoS Pathog.
9
:
e1003130
.
21.
Strickland
N.
,
T. L.
Müller
,
N.
Berkowitz
,
R.
Goliath
,
M. N.
Carrington
,
R. J.
Wilkinson
,
W. A.
Burgers
,
C.
Riou
.
2017
.
Characterization of Mycobacterium tuberculosis-specific cells using MHC class II tetramers reveals phenotypic differences related to HIV infection and tuberculosis disease.
J. Immunol.
199
:
2440
2450
.
22.
Burel
J. G.
,
C. S.
Lindestam Arlehamn
,
N.
Khan
,
G.
Seumois
,
J. A.
Greenbaum
,
R.
Taplitz
,
R. H.
Gilman
,
M.
Saito
,
P.
Vijayanand
,
A.
Sette
,
B.
Peters
.
2018
.
Transcriptomic analysis of CD4+ T cells reveals novel immune signatures of latent tuberculosis.
J. Immunol.
200
:
3283
3290
.
23.
Burel
J. G.
,
Y.
Qian
,
C.
Lindestam Arlehamn
,
D.
Weiskopf
,
J.
Zapardiel-Gonzalo
,
R.
Taplitz
,
R. H.
Gilman
,
M.
Saito
,
A. D.
de Silva
,
P.
Vijayanand
, et al
2017
.
An integrated workflow to assess technical and biological variability of cell population frequencies in human peripheral blood by flow cytometry.
J. Immunol.
198
:
1748
1758
.
24.
Lindestam Arlehamn
C. S.
,
J.
Sidney
,
R.
Henderson
,
J. A.
Greenbaum
,
E. A.
James
,
M.
Moutaftsi
,
R.
Coler
,
D. M.
McKinney
,
D.
Park
,
R.
Taplitz
, et al
2012
.
Dissecting mechanisms of immunodominance to the common tuberculosis antigens ESAT-6, CFP10, Rv2031c (hspX), Rv2654c (TB7.7), and Rv1038c (EsxJ).
J. Immunol.
188
:
5020
5031
.
25.
Arlehamn
C. L.
,
G.
Seumois
,
A.
Gerasimova
,
C.
Huang
,
Z.
Fu
,
X.
Yue
,
A.
Sette
,
P.
Vijayanand
,
B.
Peters
.
2014
.
Transcriptional profile of tuberculosis antigen-specific T cells reveals novel multifunctional features.
J. Immunol.
193
:
2931
2940
.
26.
Chen
M.
,
A.
Sutherland
,
G.
Birrueta
,
S.
Laubach
,
S.
Leonard
,
B.
Peters
,
V.
Schulten
.
2019
.
Analysis of allergen-specific T cell and IgE reactivity to different preparations of cow’s milk-containing food extracts.
Cells
8
:
667
.
27.
Birrueta
G.
,
A.
Frazier
,
A.
Pomés
,
J.
Glesner
,
S.
Filep
,
C.
Schal
,
K. Y.
Jeong
,
C.
McMurtrey
,
T.
Vander Schans
,
W. H.
Hildebrand
, et al
2019
.
Variability in German cockroach extract composition greatly impacts T cell potency in cockroach-allergic donors.
Front. Immunol.
10
:
313
.
28.
Lindestam Arlehamn
C. S.
,
D. M.
McKinney
,
C.
Carpenter
,
S.
Paul
,
V.
Rozot
,
E.
Makgotlho
,
Y.
Gregg
,
M.
van Rooyen
,
J. D.
Ernst
,
M.
Hatherill
, et al
2016
.
A quantitative analysis of complexity of human pathogen-specific CD4 T cell responses in healthy M. tuberculosis infected south africans.
PLoS Pathog.
12
:
e1005760
.
29.
Seumois
G.
,
P.
Vijayanand
,
C. J.
Eisley
,
N.
Omran
,
L.
Kalinke
,
M.
North
,
A. P.
Ganesan
,
L. J.
Simpson
,
N.
Hunkapiller
,
F.
Moltzahn
, et al
2012
.
An integrated nano-scale approach to profile miRNAs in limited clinical samples.
Am. J. Clin. Exp. Immunol.
1
:
70
89
.
30.
Picelli
S.
,
O. R.
Faridani
,
A. K.
Björklund
,
G.
Winberg
,
S.
Sagasser
,
R.
Sandberg
.
2014
.
Full-length RNA-seq from single cells using Smart-seq2.
Nat. Protoc.
9
:
171
181
.
31.
Trapnell
C.
,
L.
Pachter
,
S. L.
Salzberg
.
2009
.
TopHat: discovering splice junctions with RNA-Seq.
Bioinformatics
25
:
1105
1111
.
32.
Schmieder
R.
,
R.
Edwards
.
2011
.
Quality control and preprocessing of metagenomic datasets.
Bioinformatics
27
:
863
864
.
33.
Li
H.
,
B.
Handsaker
,
A.
Wysoker
,
T.
Fennell
,
J.
Ruan
,
N.
Homer
,
G.
Marth
,
G.
Abecasis
,
R.
Durbin
;
1000 Genome Project Data Processing Subgroup
.
2009
.
The sequence alignment/map format and SAMtools.
Bioinformatics
25
:
2078
2079
.
34.
Anders
S.
,
P. T.
Pyl
,
W.
Huber
.
2014
.
HTSeq--a Python framework to work with high-throughput sequencing data.
Bioinformatics
31
:
166
169
.
35.
Love
M. I.
,
W.
Huber
,
S.
Anders
.
2014
.
Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2.
Genome Biol.
15
:
550
.
36.
Chen
E. Y.
,
C. M.
Tan
,
Y.
Kou
,
Q.
Duan
,
Z.
Wang
,
G. V.
Meirelles
,
N. R.
Clark
,
A.
Ma’ayan
.
2013
.
Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool.
BMC Bioinformatics
14
:
128
.
37.
Kuleshov
M. V.
,
M. R.
Jones
,
A. D.
Rouillard
,
N. F.
Fernandez
,
Q.
Duan
,
Z.
Wang
,
S.
Koplev
,
S. L.
Jenkins
,
K. M.
Jagodnik
,
A.
Lachmann
, et al
2016
.
Enrichr: a comprehensive gene set enrichment analysis web server 2016 update.
Nucleic Acids Res.
44
(
W1
):
W90-7
.
38.
Robinson
J.
,
D. J.
Barker
,
X.
Georgiou
,
M. A.
Cooper
,
P.
Flicek
,
S. G. E.
Marsh
.
2020
.
IPD-IMGT/HLA database.
Nucleic Acids Res.
48
(
D1
):
D948
D955
.
39.
Du Bruyn
E.
,
S.
Ruzive
,
C. S.
Lindestam Arlehamn
,
A.
Sette
,
A.
Sher
,
D. L.
Barber
,
R. J.
Wilkinson
,
C.
Riou
.
2020
.
Mycobacterium tuberculosis-specific CD4 T cells expressing CD153 inversely associate with bacterial load and disease severity in human tuberculosis.
Mucosal Immunol.
14
:
491
499
.
40.
Dan
J. M.
,
C. S.
Lindestam Arlehamn
,
D.
Weiskopf
,
R.
da Silva Antunes
,
C.
Havenar-Daughton
,
S. M.
Reiss
,
M.
Brigger
,
M.
Bothwell
,
A.
Sette
,
S.
Crotty
.
2016
.
A cytokine-independent approach to identify antigen-specific human germinal center T follicular helper cells and rare antigen-specific CD4+ T cells in blood.
J. Immunol.
197
:
983
993
.
41.
Reiss
S.
,
A. E.
Baxter
,
K. M.
Cirelli
,
J. M.
Dan
,
A.
Morou
,
A.
Daigneault
,
N.
Brassard
,
G.
Silvestri
,
J. P.
Routy
,
C.
Havenar-Daughton
, et al
2017
.
Comparative analysis of activation induced marker (AIM) assays for sensitive identification of antigen-specific CD4 T cells.
PLoS One
12
:
e0186998
.
42.
Scholzen
T.
,
J.
Gerdes
.
2000
.
The Ki-67 protein: from the known and the unknown.
J. Cell. Physiol.
182
:
311
322
.
43.
Fletcher
H. A.
,
M. A.
Snowden
,
B.
Landry
,
W.
Rida
,
I.
Satti
,
S. A.
Harris
,
M.
Matsumiya
,
R.
Tanner
,
M. K.
O’Shea
,
V.
Dheenadhayalan
, et al
2016
.
T-cell activation is an immune correlate of risk in BCG vaccinated infants. [Published erratum appears in 2016 Nat Commun 7: 11633].
Nat. Commun.
7
:
11290
.
44.
Soares
A. P.
,
C. K.
Kwong Chung
,
T.
Choice
,
E. J.
Hughes
,
G.
Jacobs
,
E. J.
van Rensburg
,
G.
Khomba
,
M.
de Kock
,
L.
Lerumo
,
L.
Makhethe
, et al
2013
.
Longitudinal changes in CD4(+) T-cell memory responses induced by BCG vaccination of newborns.
J. Infect. Dis.
207
:
1084
1094
.
45.
Adekambi
T.
,
C. C.
Ibegbu
,
S.
Cagle
,
A. S.
Kalokhe
,
Y. F.
Wang
,
Y.
Hu
,
C. L.
Day
,
S. M.
Ray
,
J.
Rengarajan
.
2015
.
Biomarkers on patient T cells diagnose active tuberculosis and monitor treatment response.
J. Clin. Invest.
125
:
1827
1838
.
46.
Riou
C.
,
N.
Berkowitz
,
R.
Goliath
,
W. A.
Burgers
,
R. J.
Wilkinson
.
2017
.
Analysis of the phenotype of Mycobacterium tuberculosis-specific CD4+ T cells to discriminate latent from active tuberculosis in HIV-uninfected and HIV-infected individuals.
Front. Immunol.
8
:
968
.
47.
Riou
C.
,
E.
Du Bruyn
,
S.
Ruzive
,
R. T.
Goliath
,
C. S.
Lindestam Arlehamn
,
A.
Sette
,
A.
Sher
,
D. L.
Barber
,
R. J.
Wilkinson
.
2020
.
Disease extent and anti-tubercular treatment response correlates with Mycobacterium tuberculosis-specific CD4 T-cell phenotype regardless of HIV-1 status.
Clin. Transl. Immunology
9
:
e1176
.
48.
Musvosvi
M.
,
D.
Duffy
,
E.
Filander
,
H.
Africa
,
S.
Mabwe
,
L.
Jaxa
,
N.
Bilek
,
A.
Llibre
,
V.
Rouilly
,
M.
Hatherill
, et al
2018
.
T-cell biomarkers for diagnosis of tuberculosis: candidate evaluation by a simple whole blood assay for clinical translation.
Eur. Respir. J.
51
:
1800153
.
49.
Silveira-Mattos
P. S.
,
B.
Barreto-Duarte
,
B.
Vasconcelos
,
K. F.
Fukutani
,
C. L.
Vinhaes
,
D.
Oliveira-De-Souza
,
C. C.
Ibegbu
,
M. C.
Figueiredo
,
T. R.
Sterling
,
J.
Rengarajan
,
B. B.
Andrade
.
2020
.
differential expression of activation markers by Mycobacterium tuberculosis-specific CD4+ T cell distinguishes extrapulmonary from pulmonary tuberculosis and latent infection.
Clin. Infect. Dis.
71
:
1905
1911
.
50.
Riou
C.
,
C. M.
Gray
,
M.
Lugongolo
,
T.
Gwala
,
A.
Kiravu
,
P.
Deniso
,
L.
Stewart-Isherwood
,
S. V.
Omar
,
M. P.
Grobusch
,
G.
Coetzee
, et al
2014
.
A subset of circulating blood mycobacteria-specific CD4 T cells can predict the time to Mycobacterium tuberculosis sputum culture conversion.
PLoS One
9
:
e102178
.
51.
Ahmed
M. I. M.
,
N. E.
Ntinginya
,
G.
Kibiki
,
B. A.
Mtafya
,
H.
Semvua
,
S.
Mpagama
,
C.
Mtabho
,
E.
Saathoff
,
K.
Held
,
R.
Loose
, et al
2018
.
Phenotypic changes on Mycobacterium Tuberculosis-specific CD4 T Cells as surrogate markers for tuberculosis treatment efficacy.
Front. Immunol.
9
:
2247
.
52.
Canaday
D. H.
,
R. J.
Wilkinson
,
Q.
Li
,
C. V.
Harding
,
R. F.
Silver
,
W. H.
Boom
.
2001
.
CD4(+) and CD8(+) T cells kill intracellular Mycobacterium tuberculosis by a perforin and Fas/Fas ligand-independent mechanism.
J. Immunol.
167
:
2734
2742
.
53.
Bastian
M.
,
T.
Braun
,
H.
Bruns
,
M.
Röllinghoff
,
S.
Stenger
.
2008
.
Mycobacterial lipopeptides elicit CD4+ CTLs in Mycobacterium tuberculosis-infected humans.
J. Immunol.
180
:
3436
3446
.
54.
Commandeur
S.
,
M.
Coppola
,
K.
Dijkman
,
A. H.
Friggen
,
K. E.
van Meijgaarden
,
S. J.
van den Eeden
,
L.
Wilson
,
J. J.
van der Ploeg-van Schip
,
K. L.
Franken
,
A.
Geluk
,
T. H.
Ottenhoff
.
2014
.
Clonal analysis of the T-cell response to in vivo expressed Mycobacterium tuberculosis protein Rv2034, using a CD154 expression based T-cell cloning method.
PLoS One
9
:
e99203
.
55.
Harari
A.
,
V.
Rozot
,
F.
Bellutti Enders
,
M.
Perreau
,
J. M.
Stalder
,
L. P.
Nicod
,
M.
Cavassini
,
T.
Calandra
,
C. L.
Blanchet
,
K.
Jaton
, et al
2011
.
Dominant TNF-α+ Mycobacterium tuberculosis-specific CD4+ T cell responses discriminate between latent infection and active disease.
Nat. Med.
17
:
372
376
.
56.
Chávez-Galán
L.
,
M. C.
Arenas-Del Angel
,
E.
Zenteno
,
R.
Chávez
,
R.
Lascurain
.
2009
.
Cell death mechanisms induced by cytotoxic lymphocytes.
Cell. Mol. Immunol.
6
:
15
25
.
57.
Ahmed
A.
,
V.
Adiga
,
S.
Nayak
,
J. A. J.
Uday Kumar
,
C.
Dhar
,
P. N.
Sahoo
,
B. K.
Sundararaj
,
G. D.
Souza
,
A.
Vyakarnam
.
2018
.
Circulating HLA-DR+CD4+ effector memory T cells resistant to CCR5 and PD-L1 mediated suppression compromise regulatory T cell function in tuberculosis.
PLoS Pathog.
14
:
e1007289
.
58.
Yang
Q.
,
M.
Zhang
,
Q.
Chen
,
W.
Chen
,
C.
Wei
,
K.
Qiao
,
T.
Ye
,
G.
Deng
,
J.
Li
,
J.
Zhu
, et al
2020
.
Cutting Edge: Characterization of human tissue-resident memory T cells at different infection sites in patients with tuberculosis.
J. Immunol.
204
:
2331
2336
.
59.
Tian
Y.
,
G.
Seumois
,
L. M.
De-Oliveira-Pinto
,
J.
Mateus
,
S.
Herrera-de la Mata
,
C.
Kim
,
D.
Hinz
,
N. D. S.
Goonawardhana
,
A. D.
de Silva
,
S.
Premawansa
, et al
2019
.
Molecular signatures of dengue virus-specific IL-10/IFN-γ co-producing CD4 T cells and their association with dengue disease.
Cell Rep.
29
:
4482
4495.e4
.
60.
DeGrendele
H. C.
,
M.
Kosfiszer
,
P.
Estess
,
M. H.
Siegelman
.
1997
.
CD44 activation and associated primary adhesion is inducible via T cell receptor stimulation.
J. Immunol.
159
:
2549
2553
.
61.
Baaten
B. J.
,
R.
Tinoco
,
A. T.
Chen
,
L. M.
Bradley
.
2012
.
Regulation of antigen-experienced T cells: Lessons from the quintessential memory marker CD44.
Front. Immunol.
3
:
23
.
62.
Callan
M. F.
,
L.
Tan
,
N.
Annels
,
G. S.
Ogg
,
J. D.
Wilson
,
C. A.
O’Callaghan
,
N.
Steven
,
A. J.
McMichael
,
A. B.
Rickinson
.
1998
.
Direct visualization of antigen-specific CD8+ T cells during the primary immune response to Epstein-Barr virus In vivo.
J. Exp. Med.
187
:
1395
1402
.
63.
Doisne
J. M.
,
A.
Urrutia
,
C.
Lacabaratz-Porret
,
C.
Goujard
,
L.
Meyer
,
M. L.
Chaix
,
M.
Sinet
,
A.
Venet
.
2004
.
CD8+ T cells specific for EBV, cytomegalovirus, and influenza virus are activated during primary HIV infection.
J. Immunol.
173
:
2410
2418
.
64.
Chandele
A.
,
J.
Sewatanon
,
S.
Gunisetty
,
M.
Singla
,
N.
Onlamoon
,
R. S.
Akondy
,
H. T.
Kissick
,
K.
Nayak
,
E. S.
Reddy
,
H.
Kalam
, et al
2016
.
Characterization of human CD8 T cell responses in dengue virus-infected patients from india.
J. Virol.
90
:
11259
11278
.
65.
Miller
J. D.
,
R. G.
van der Most
,
R. S.
Akondy
,
J. T.
Glidewell
,
S.
Albott
,
D.
Masopust
,
K.
Murali-Krishna
,
P. L.
Mahar
,
S.
Edupuganti
,
S.
Lalor
, et al
2008
.
Human effector and memory CD8+ T cell responses to smallpox and yellow fever vaccines.
Immunity
28
:
710
722
.
66.
Schwenk
R.
,
G.
Banania
,
J.
Epstein
,
Y.
Kim
,
B.
Peters
,
M.
Belmonte
,
H.
Ganeshan
,
J.
Huang
,
S.
Reyes
,
A.
Stryhn
, et al
2013
.
Ex vivo tetramer staining and cell surface phenotyping for early activation markers CD38 and HLA-DR to enumerate and characterize malaria antigen-specific CD8+ T-cells induced in human volunteers immunized with a Plasmodium falciparum adenovirus-vectored malaria vaccine expressing AMA1.
Malar. J.
12
:
376
.
67.
Thevarajan
I.
,
T. H. O.
Nguyen
,
M.
Koutsakos
,
J.
Druce
,
L.
Caly
,
C. E.
van de Sandt
,
X.
Jia
,
S.
Nicholson
,
M.
Catton
,
B.
Cowie
, et al
2020
.
Breadth of concomitant immune responses prior to patient recovery: a case report of non-severe COVID-19.
Nat. Med.
26
:
453
455
.
68.
Wang
F.
,
H.
Hou
,
Y.
Luo
,
G.
Tang
,
S.
Wu
,
M.
Huang
,
W.
Liu
,
Y.
Zhu
,
Q.
Lin
,
L.
Mao
, et al
2020
.
The laboratory tests and host immunity of COVID-19 patients with different severity of illness.
JCI Insight
5
:
e137799
.
69.
Mathew
D.
,
J. R.
Giles
,
A. E.
Baxter
,
D. A.
Oldridge
,
A. R.
Greenplate
,
J. E.
Wu
,
C.
Alanio
,
L.
Kuri-Cervantes
,
M. B.
Pampena
,
K.
D’Andrea
, et al
UPenn COVID Processing Unit
.
2020
.
Deep immune profiling of COVID-19 patients reveals distinct immunotypes with therapeutic implications.
Science
369
:
eabc8511
.
70.
Kuri-Cervantes
L.
,
M. B.
Pampena
,
W.
Meng
,
A. M.
Rosenfeld
,
C. A. G.
Ittner
,
A. R.
Weisman
,
R. S.
Agyekum
,
D.
Mathew
,
A. E.
Baxter
,
L. A.
Vella
, et al
2020
.
Comprehensive mapping of immune perturbations associated with severe COVID-19.
Sci. Immunol.
5
:
eabd7114
.
71.
Braun
J.
,
L.
Loyal
,
M.
Frentsch
,
D.
Wendisch
,
P.
Georg
,
F.
Kurth
,
S.
Hippenstiel
,
M.
Dingeldey
,
B.
Kruse
,
F.
Fauchere
, et al
2020
.
SARS-CoV-2-reactive T cells in healthy donors and patients with COVID-19.
Nature
587
:
270
274
.
72.
Meckiff
B. J.
,
C.
Ramírez-Suástegui
,
V.
Fajardo
,
S. J.
Chee
,
A.
Kusnadi
,
H.
Simon
,
S.
Eschweiler
,
A.
Grifoni
,
E.
Pelosi
,
D.
Weiskopf
, et al
2020
.
Imbalance of regulatory and cytotoxic SARS-CoV-2-reactive CD4+ T Cells in COVID-19.
Cell
183
:
1340
1353.e16
.

The authors have no financial conflicts of interest.

Supplementary data